1 INFO-VAX	Tue, 24 Oct 2000	Volume 2000 : Issue 595       Contents:0 Re: AlphaServer 2100 upgrade from 4/233 to 5/2500 Re: AlphaServer 2100 upgrade from 4/233 to 5/250- Captain RTR's Syphilitic Hoard of Cut-Throats 1 Re: Captain RTR's Syphilitic Hoard of Cut-Throats  Compaq & Candle  Compuware will kill OpenVMS ??? # Re: Compuware will kill OpenVMS ???  Copy Users to Another System  Re: Copy Users to Another System DECdtm QUESTIONS Re: DECdtm QUESTIONS Re: DECdtm QUESTIONS Re: DECdtm QUESTIONS) Determining if timer request ID valid (?) - Re: Determining if timer request ID valid (?)  eBay (guess what) again. Re: eBay (guess what) again. Re: FMS and GKS ( Re: FS microvax 3100-85, 128mb, 6gb, 8mm Future of VT, etc . . .  Re: Galaxy doubt Re: Galaxy doubt Re: Galaxy doubt (now way OT) 
 GnuPG status? & Re: HP4050 vs HP4050TN when using DCPS" Inexpensive Laser Printers for VMS RE: Java on Openvms 3 Re: Linker performance on DS20 slower than VAX 6520 , Re: loss of linefeeds when typing a files-11$ mail$system_flags, DECNET IV & alias( Re: mail$system_flags, DECNET IV & alias Need VMS on tape cartridge Re: Odd File Expiration Dates  Re: Odd File Expiration Dates   Re: OpenVMS UK (London) Roadtrip  Re: OpenVMS UK (London) Roadtrip Oracle 7 and 8i on same box? Oracle 7 and 8i on same box? Re: PDF under OpenVMS  Re: PDF under OpenVMS  Re: PDF under OpenVMS  Re: PDF under OpenVMS  Re: PDF under OpenVMS  Re: PDF under OpenVMS  Re: PDF under OpenVMS ' Re: Press Release from Process Software : Protect yourself and your computer get Evidence Eliminator Re: Q: Why not (2^n)-bit?  Re: Q: Why not (2^n)-bit?  Re: Q: Why not (2^n)-bit? 5 RE: Question about running out of queue entry numbers  Re: Record not current- Re: RSH Problem - Can someone try this for me  Shared data , Re: splitting infinitives (was Galaxy doubt)" Re: Stockholm Exchange OM and VMS?" Re: Stockholm Exchange OM and VMS?, Re: SUMMARY: installing VMS 7.2-1 on 433au ? Re: TCP/IP and load balancing  RE: TCP/IP and load balancing  RE: TCP/IP and load balancing / Re: Trade-offs regarding Installed Known Images  Re: VAX to Alpha (DS10) ! Re: Very long filenames in VMSTAR  Re: video card info at Re: VMS V7.2-1 Re: VMS V7.2-1 Re: vt520 music question Re: vt520 music question Re: What does DCPS really do ?C Re: What would be your ultimate dream VMS machine (desktop/tower) ? C Re: What would be your ultimate dream VMS machine (desktop/tower) ? C Re: What would be your ultimate dream VMS machine (desktop/tower) ? C Re: What would be your ultimate dream VMS machine (desktop/tower) ? C Re: What would be your ultimate dream VMS machine (desktop/tower) ? ; why just single-user licenses with the educational package? ? Re: why just single-user licenses with the educational package?   F ----------------------------------------------------------------------   Date: 24 Oct 2000 11:12 -0400  From: hein@eps.zko.dec.c*m9 Subject: Re: AlphaServer 2100 upgrade from 4/233 to 5/250 & Message-ID: <24OCT200011122436@miasys>  W In article <sv9cipob0j916d@corp.supernews.com>, "Mark E. Levy" <mark@fsi.net> writes... I >I'm upgrading an Alphaserver 2100 from two 4/233 CPUs to two 5/250 CPUs. K >I'll probably have to change the firmware from the EV4 to the EV5 version.  > - >Is there anything else I need to know about?   7 Yeah, you need to check out the firmware release notes. N Included some starting data from my archive, but check your own firmware CDs'.  	 -------   N   Attached below are the current V4.1 release notes to be included in the nextN firmware CD release (Aug. '95).  There are apendices that cover the upgrade orH downgrade of ev4 to ev5 and back to ev4.  Also listed in those notes areJ minimum requirements for the EISA Configuration Utility (ECU) and the RAIDI Standalone Utility for EV5 systems.  ECU is V1.8 and the RAID firmware is K V2.15 for EIAS based and V2.19 for PCI based.  Kit info. on these is in the 3 notes.  Minimum O.S. levels listed for EV5 systems:  		Digital UNIX 3.2B  		OpenVMS V6.2 		Microsoft Windows NT V3.51& The bootfiles found on the v3.2 FW CD: 	for EV4 system update:  		[alpha2100]as2100_e4_v4_0.exe  	for EV5 system update:  		[alpha2100]as2100_e5_v4_0.exe  	for EV4 to EV5 system upgrade:  		[alpha2100]v40_sbupgrade.exe  :  *********************************************************:  * UPGRADE 4/200 or 4/275 to 5/250 (EV4 or EV4.5 to EV5) *:  *********************************************************  " Basic steps to follow for upgrade:
 	Boot the CD:  		P00>>> b dka600 -flags 0,a0 ' 		Bootfile:[alpha2100]v40_sbupgrade.exe ' 	Update the ARC and SRM console images:  		UPD> update *arc*  		UPD> update *srm*  		UPD> exit - 	Power off swap out EV4 CPUs swap in EV5 CPUs 	 	Power up O Note: I mentioned running update at this point but since I suggest not updating O the Fail Safe Loader (FSL) this update step is not needed.  The assumption here O is that the customer will be downgrding to the EV4 after running the EV5 bench- 
 mark testing.  	Run ECU( 	Run RAID Standalone Utility (if needed)  <  ***********************************************************<  * DOWNGRADE 5/250 to 4/200 or 4/275 (EV5 to EV4 or EV4.5) *<  ***********************************************************  O Downgrade info. is the release notes...it is t a manual update procedure using   the EV5 update file.  O   The steps will force the ev4 SRM and ARC code to be written to the flashroms. K You must first blast in the ev4 code while the ev5 CPU is in the system and 5 after that you can then swap the cpu from ev5 to ev4.   4 Basic Steps to follow for downgrade from EV5 to EV4:
 	Boot the CD:  		P00>>> b dka600 -flags 0,a0 ( 		Bootfile:[alpha2100]as2100_e5_v4_0.exe1 	Manual Update of the ARC and SRM console images:  		UPD> exit . 		Do you want to do a manual update? [y/(n)] YO Note: the -i qualifier in these next two steps are to force the update to occur 4 		P00>>>update -path ev4srmrom -target sbsrmflash -i4 		P00>>>update -path ev4arcrom -target sbarcflash -i 		P00>>>lfu 
 		UPD>exit- 	Power off swap out EV5 CPUs swap in EV4 CPUs M Because the fail safe loader was not changed you shuold not have to rerun the  EV4 update procedure. 	 	Power up  	Run ECU* 	Run RAID standalone Utility if necessary.    - ------------- an other note i found ---------   "     To clarify a couple of points:  J     It is not really necessary to run ECU or RCU each time when repeatedlyJ     swapping between EV4 and EV5 CPUs. It is included in the 5/250 upgradeH     document only to make sure that ECU V1.8 is run once; the assumptionH     was that customers might be upgrading older systems on which ECU hadK     not been run recently. Once ECU V1.8 has been run with either an EV4 or D     EV5 processor, it does not need to be run again (unless the EISA&     configuration changes, of course).  K     Also, if you keep the EV4 Fail Safe Loader in your FLASH ROMs, then you F     can downgrade just by swapping CPUs and using the FSL to boot yourK     console, then boot your operating system. This would only be applicable J     if you're downgrading for a quick experiment and you're not running NTE     and therefore don't need ARC for your test. Keep in mind that you I     wouldn't want to run this way indefinitely; if you lose power for any I     reason, you'll come back up to the fail safe loader and won't be able      to boot automatically.  $ ------------------------------------   hth, 	Hein.   ------------------------------   Date: 24 Oct 2000 15:50:01 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)9 Subject: Re: AlphaServer 2100 upgrade from 4/233 to 5/250 6 Message-ID: <8t4b39$92q$1@mailint03.im.hou.compaq.com>  g In article <39F4DD94.13ECB6E7@earthlink.net>, "David J. Dachtera" <djesys.nospam@earthlink.net> writes: " :Well, actually it's a Tru64 box.   #   Sorry, wrong newsgroup then.  :-)   $ :On power up, the box hangs with theH :front panel display reading "Starting CPU 0". We were wondering if thisG :is a firmware issue (EV4 Vs. EV5) or what we might be looking at here.   F   Could well be -- you do need to load new firmware and a new failsafeE   loader (FSL) when upgrading or downgrading between EV4 and EV5 CPUs    on this series of platforms.  G   See the details on upgrading and downgrading and loading new firmware    and a new FSL at:   =   ftp://ftp.digital.com/pub/Digital/Alpha/firmware/index.html F   ftp://ftp.digital.com/pub/Digital/Alpha/firmware/readmes/as2x00.html  H   To locate the FSL jumper or FSL switch, based on the particular model:  5   Enable the FSL as follows depending on system type: I     AlphaServer 2000: Install jumper J6 on the AlphaServer 2000 Backplane B     AlphaServer 2100: Install jumper J6 on the Standard IO module.F     AlphaServer 2100A LP: Install jumper J6 on the Standard IO module.G     AlphaServer 2100A: Move the FSL enable switch -- the top switch on  H                        the backplane module -- rightwards to enable FSL.H     AlphaServer 2100A RM: Move the FSL enable switch -- the right switchJ                        on the backplane module -- downwards to enable FSL.  I     On the AlphaServer 2x00 series standard I/O module, J6 is located to  4     the right of the row of ribbon cable connectors.  K     On the AlphaServer 2000 series, with the EISA slots oriented downwards  J     and to the left, J6 is the uppermost of the three jumpers located off -     the right end of the array of EISA slots.   N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  # Date: Tue, 24 Oct 2000 13:53:15 GMT  From: richard_maher@my-deja.com 6 Subject: Captain RTR's Syphilitic Hoard of Cut-Throats) Message-ID: <8t4487$389$1@nnrp1.deja.com>    Hi,   C I'm obviously being as articulate as a not very articulate thing. I F agree with everything you've just written but what *you* don't seem toD realize is that VMS *already* has a standard transaction API and hasE had for ten years. It's called DECdtm! Unfortunately to date, most of G the DECdtm API has been concealed from customers by petty little empire C builders and their cynical attempts at extortion. Also, please note F that unlike an RTL the DECdtm API consists of system services that canC be called from inner modes for those resource managers and customer F applications that may require it. Critically, DECdtm is also fast! ButF this speed will be jeopardized if RTR has to write txn/url info to its1 journal in addition to the prevailing DECdtm i/o.   F What Compaq needs to do now is redress the atrophy inflicted on DECdtmG in its wilderness years! Like it or not VMS is currently playing catch- D up to Microsoft and Tandem with eBusiness and internet transactions.E The Transaction Internet Protocol world is upon us. It's time for VMS D to cross the Rubicon or perish. What we should be talking about hereC is "How can UCX IPv6 help our DECdtm/TIP implementation achieve TLS D (Transport Layer Security)?" or "DECdtm's new multiplexing algorithmG has given us an X tps increase in performance!" Instead all efforts are C being expended trying to repel captain RTR's syphilitic hoards from  boarding the good ship TIP :-(  E Now, a lot of you seem to get your jollies slagging off Microsoft (Oh F did I laugh the 1000th time I read Micro$hit) but remember this, it isE VMS management that is forcing developers to buy/use RTR if they want D to get TIP, whereas Microsoft has said "We strongly suggest that youD use COM+ but if you don't want to, and you'd like to choose your ownE 2nd pipe then here's DtcGetTransactionManagerEx et al - Have fun with D our URLs." What is even worse (inconceivable) is that VMS managementE seriously expects customers to install RTR on *every* server involved G in a transaction - Disbelief! Let me see - I'm running Windows 2000, do 0 I use a) MTS or b) RTR? Oh, the inner turmoil!!!  F One day soon an XML page, a COM+ object invocation or a Socket messageE is going to arrive at your VMS server's doorstep with an embedded TIP ) URL. What are you going to do about it???    Regards Richard Maher     F >I see your point of view. However, to be the devil's advocate, havingF >a single standard API to ensure transcation routing on VMS will spurrD >development of applications that use this API. And knowing that allC >VMS sites have that API available means that applications that are G >written to use will run on any VMS box without forcing the customer to  >buy another piece of software.   D >This is similar in principle, in my opinion, to providing the C andF >other compiler RTLs with the VMS system instead of with the compiler.    & Sent via Deja.com http://www.deja.com/ Before you buy.    ------------------------------  # Date: Tue, 24 Oct 2000 16:55:34 GMT ? From: Jim.Johnson@software-exploration.nospam.com (Jim Johnson) : Subject: Re: Captain RTR's Syphilitic Hoard of Cut-Throats0 Message-ID: <39f5baae.11985944@news.demon.co.uk>   Richard,  @ First: calm down. Right now the quality of argument that you are> making is not going to help convince anyone that you're right.   Now:  E I've been working on setting up a meeting with VMS in the coming week E to discuss what their options are, and how/if I can help.  Until I've B had a chance to talk to the people actually involved, I personally? don't want to jump to any conclusions about what they're doing.   D I will report back on what details of any outcome that I feel I can,# but right now I can't promise much.     ? What you (all of you, actually) can do is to make the case that ? investment in TP on VMS is worth making.  Do any of you develop D software on VMS still?  If you do, well I'd be interested in hearingC what you'd like to see in this area, but much more importantly, you  NEED TO TELL VMS DEVELOPMENT.   F Let me play devil's advocate on this for a second.  There's been a lotE of sturm und drang on this topic in the last few weeks.  However, the D number of people who have shown sufficient interest in this topic to= enter a note have been outweighed by the number who have been + concerned about the latest port of nethack.   E If I was using this forum as a basis for how important this topic is, @ I'd conclude that if I can get something that sort of looks likeE something that could be called a solution, and if it wouldn't cost me E much of anything to do it, then maybe it would be worth doing just to F quiet down a few people.  What I wouldn't conclude from the discussionF here is that this is an area that VMS should work to be leadership (or even a leading competitor) in.  = Personally, I do think this is a very important area for VMS, F especially if it plans on any sort of back room role.  However, as I'mB not a major paying customer, that case isn't one that I can easily make.    Jim.  F PS: If we want to talk technicalities for a second, there are indeed aD couple of standards that VMS should offer -- this is what the XA andC TIP discussions are about.  Right now neither DECdtm nor RTR supply C these as supported features.  Either one would have to change to do F so.  My original claim, and one that I stick to, is that the effort to do this via DECdtm is minimal. Jim Johnson  Software Exploration, Ltd.' Software Navigation and Discovery Tools    ------------------------------  % Date: Tue, 24 Oct 2000 15:57:05 -0200 ) From: fabio_compaq@ep-bc.petrobras.com.br  Subject: Compaq & CandleL Message-ID: <OFF1A46A20.CF707503-ON83256982.00628231@ep-bc.petrobras.com.br>   People   Check the link  W http://www.candle.com/news_events/press_releases/consulting_services/compaq_080299.html    Regards    FC.    ------------------------------  % Date: Tue, 24 Oct 2000 16:24:20 +0200 ! From: Luminy <luminyGG@iname.com> ( Subject: Compuware will kill OpenVMS ???9 Message-ID: <MPG.145fb9cc72066c0d989683@news2.hexanet.fr>    Hi,   C In June 99, Rich Marcello told us that Compuware was partnership of " OpenVMS, but they didn'it confirm.  ( In September 99 they confirmed, but.....  < Last week, i saw  a prsentation of Uniface 8 and i was glad& when i put this project on my VMS/RDB.5 But, unfortunatly, after the meeting, they said to me ' that platform OpenVMS is out of scope !   F What's about the partnership between Compaq and Compuware on OpenVMS ?   A Dopu   -- C	 Luminy...e1 "Il n'y a qu'une vrit, mais chacun la sienne !"    ------------------------------  # Date: Tue, 24 Oct 2000 15:47:43 GMTa% From: A.Greig@virgin.net (Alan Greig)y, Subject: Re: Compuware will kill OpenVMS ???2 Message-ID: <39f5adc5.1298762912@news.newsguy.com>  F On Tue, 24 Oct 2000 16:24:20 +0200, Luminy <luminyGG@iname.com> wrote:   >Hi, >mD >In June 99, Rich Marcello told us that Compuware was partnership of# >OpenVMS, but they didn'it confirm.r >e) >In September 99 they confirmed, but.....r >e= >Last week, i saw  a prsentation of Uniface 8 and i was gladt' >when i put this project on my VMS/RDB.a6 >But, unfortunatly, after the meeting, they said to me( >that platform OpenVMS is out of scope ! >dG >What's about the partnership between Compaq and Compuware on OpenVMS ?S  D Rich Marcello is aware that employees of some partners are repeatingC false information about VMS support out of ignorance. This could betA the case here. A search of www.compuware.com  for OpenVMS finds as> document dated 5th October 2000 which lists VMS as a supportedC platform for Uniface. There are no references to de-supporting VMS.   ? I'd suggest you email Richard.Marcello@compaq.com directly withdD details. He really does want to hear directly of any stories such as these.   --
 Alan Greig   ------------------------------  % Date: Tue, 24 Oct 2000 11:20:15 +0100 ( From: "Vitor Sarabando" <vss@liscont.pt>% Subject: Copy Users to Another Systemr3 Message-ID: <lodJ5.369$k412.589981@news.jazztel.pt>l   Hi...   @ I'm Trying to copy users from one system to another, does anyone: knows how it's done or do i have to create them by scratch   Thanks   Vitor Sarabandor vss@liscont.pt   ------------------------------  # Date: Tue, 24 Oct 2000 14:01:03 GMTi/ From: Mike Price <mike.price@littlewoods.co.uk> ) Subject: Re: Copy Users to Another System?) Message-ID: <8t44mq$3nh$1@nnrp1.deja.com>u  3 In article <lodJ5.369$k412.589981@news.jazztel.pt>, +   "Vitor Sarabando" <vss@liscont.pt> wrote:e > Hi...f >rB > I'm Trying to copy users from one system to another, does anyone< > knows how it's done or do i have to create them by scratch >1 > Thanks >2 > Vitor Sarabando  > vss@liscont.pt  E I have done this by copying the records from the sysuaf file from oneuF to the other. The format of the sysuaf files is documented (I believe)D but it is easy enought to access it from DCL like any normal indexed file  D There are also uaf calls you can do from a program if you want to go that way  D However, last time I had to do this (last week actually) I had aboutE 500 users to move which all did the same thing. I simply created somenF very simple DCL that read a file containing the required userids (fromA sysuaf list) and then used authorize to create the new userids by- copying from a default account.    DCL was something like  
 $ set noon $ say:==write sys$output $ cls:==typ/p nl:4 $ menu:/ $ open f1 sysuaf.current/read2 $ loop:y $ read f1 rec/end=eofp $ rec = f$edit(rec,"compress")
 $ goto add $ eof:
 $ close f1 $ exit $! $ add: $! $ userid = f$element(1," ",rec) 7 $ owner = f$element(7," ",rec)+" "+f$element(8," ",rec)h $ passw: $ password = "newpassword"& $ say "Userid "+userid+" owner "+owner $ owner = """"+owner+"""" & $ mc authorize copy/noadd user_default/ 'userid'/passw='password'/owner='owner'/&#9670;r $ goto loope $!     sysuaf.current format is  @ User xxxxx       last login date  13-OCT-2000  Owner Fred Bloggs? User xxxxx       last login date  5-MAR-2000  Owner Fred Bloggse             Either way seems to work   HTHf   Mike >  >  --B All opinions are my own and do not necessarily reflect those of my employer    & Sent via Deja.com http://www.deja.com/ Before you buy.r   ------------------------------  % Date: Tue, 24 Oct 2000 12:06:28 -0200a) From: fabio_compaq@ep-bc.petrobras.com.br  Subject: DECdtm QUESTIONS L Message-ID: <OFF9F73828.74A9B449-ON83256982.004D3B5C@ep-bc.petrobras.com.br>  K Do the DECdtm works over SCS  in a Cluster Configuration ? What means, if Is begin to use SMCI (SharedaF Memory Cluster Interconnect) , it will  improve the speed of my DECdtm transactions ???   Regards,   FC.e   ------------------------------  # Date: Tue, 24 Oct 2000 14:31:40 GMT* From: richard_maher@my-deja.comE Subject: Re: DECdtm QUESTIONSr) Message-ID: <8t46gd$5bg$1@nnrp1.deja.com>*   Hi,*  G If SMCI is Memory Channel then my best guess is YES it will improve theDF speed of your DECdtm transactions. (Although everyone seems to suggest@ that MC is better suited to LARGE data transfers compared to the! miniscule requirements of DECdtm)i  E My understanding is that DECnet knows to use the fastest interconnecteC available and if that is SMCI then that's what it, and consequently E DECdtm, will use. Someone who knows DECnet internals would be of morep use to you.s  F Obviously the TCP/IP DECdtm code is begging to be released but whetherD the TCP/IP code is Memory Channel aware, I don't know. Anyway all ofF the unpublished DECdtm functionality is becoming too painful for me toF talk about but God bless you and keep you for actually doing something with DECdtm!   Regards Richard Maher.  7 In article <OFF9F73828.74A9B449-ON83256982.004D3B5C@ep-m bc.petrobras.com.br>,u,   fabio_compaq@ep-bc.petrobras.com.br wrote:A > Do the DECdtm works over SCS  in a Cluster Configuration ? What  means, if It > begin to use SMCI (SharedtH > Memory Cluster Interconnect) , it will  improve the speed of my DECdtm > transactions ??? > 
 > Regards, >  > FC.n >d >r    & Sent via Deja.com http://www.deja.com/ Before you buy.    ------------------------------  % Date: Tue, 24 Oct 2000 09:59:21 -0500R+ From: "Mike Kier" <michael.kier@compaq.com>n Subject: Re: DECdtm QUESTIONS]6 Message-ID: <8t483s$7j2$1@mailint03.im.hou.compaq.com>  L SMCI is not memory channel, rather it is code implementing the SCS transportL utilizing shared memory configured between VMS instances that are members of+ the same VMScluster within a Galaxy system.4  L The data transported via SMCI does not leave the confines of the system box.   Mike  , <richard_maher@my-deja.com> wrote in message# news:8t46gd$5bg$1@nnrp1.deja.com...r > Hi,  >-I > If SMCI is Memory Channel then my best guess is YES it will improve theiH > speed of your DECdtm transactions. (Although everyone seems to suggestB > that MC is better suited to LARGE data transfers compared to the# > miniscule requirements of DECdtm)o >kG > My understanding is that DECnet knows to use the fastest interconnectmE > available and if that is SMCI then that's what it, and consequentlynG > DECdtm, will use. Someone who knows DECnet internals would be of more,
 > use to you.e >oH > Obviously the TCP/IP DECdtm code is begging to be released but whetherF > the TCP/IP code is Memory Channel aware, I don't know. Anyway all ofH > the unpublished DECdtm functionality is becoming too painful for me toH > talk about but God bless you and keep you for actually doing something > with DECdtm! >r > Regards Richard Maher. >e9 > In article <OFF9F73828.74A9B449-ON83256982.004D3B5C@ep-e > bc.petrobras.com.br>,p. >   fabio_compaq@ep-bc.petrobras.com.br wrote:C > > Do the DECdtm works over SCS  in a Cluster Configuration ? Whatm
 > means, if Ie > > begin to use SMCI (SharedaJ > > Memory Cluster Interconnect) , it will  improve the speed of my DECdtm > > transactions ??? > >- > > Regards, > >  > > FC.4 > >2 > >5 >: >G( > Sent via Deja.com http://www.deja.com/ > Before you buy.c   ------------------------------  # Date: Tue, 24 Oct 2000 16:26:56 GMT$? From: Jim.Johnson@software-exploration.nospam.com (Jim Johnson)E Subject: Re: DECdtm QUESTIONSh0 Message-ID: <39f5b7ff.11298266@news.demon.co.uk>  E If a DECdtm instance is attempting to talk to another DECdtm instance E located elsewhere in the same cluster, it will use SCS (via IPC).  Itr@ will not go to DECnet, unless the IPC services have been changed	 severely.    Jim.  # On Tue, 24 Oct 2000 12:06:28 -0200,e* fabio_compaq@ep-bc.petrobras.com.br wrote:  L >Do the DECdtm works over SCS  in a Cluster Configuration ? What means, if I >begin to use SMCI (SharedG >Memory Cluster Interconnect) , it will  improve the speed of my DECdtmg >transactions ???n > 	 >Regards,  >  >FC. >t >i   Jim Johnsoni Software Exploration, Ltd.' Software Navigation and Discovery Tools/   ------------------------------   Date: 24 Oct 2000 16:04:09 GMT6 From: DAVISM@er6.eng.ohio-state.edu (Michael T. Davis)2 Subject: Determining if timer request ID valid (?): Message-ID: <8t4btp$qd4$1@charm.magnus.acs.ohio-state.edu>  D 	Is there a way to determine if there are any timer requests pendingI given a timer request ID?  For example, say I'm implementing a routine to.K be called after mapping into memory via LIB$FIND_IMAGE_SYMBOL.  The routine N needs to use a timer to determine if/when a timeout condition occurs.  There'sL no guarantee that the calling routine/image isn't using any timers.  How canN my routine set a timer with a unique timer request ID, such that if cancelled,: it won't effect any timers from the calling routine/image?   Thanks,i  Miket --K              Michael T. Davis              |    Systems Specialist: ChE,MSEeN   E-mail: davism@er6.eng.ohio-state.edu    | Departmental Networking/ComputingJ            -or- DAVISM+@osu.edu            |     The Ohio State UniversityJ http://www.er6.eng.ohio-state.edu/~davism/ |     197 Watts, (614) 292-6928   ------------------------------  # Date: Tue, 24 Oct 2000 16:51:50 GMTz= From: system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-)c6 Subject: Re: Determining if timer request ID valid (?)0 Message-ID: <009F2141.A3DAFFF2@SendSpamHere.ORG>  s In article <8t4btp$qd4$1@charm.magnus.acs.ohio-state.edu>, DAVISM@er6.eng.ohio-state.edu (Michael T. Davis) writes:t >gE >	Is there a way to determine if there are any timer requests pendingeJ >given a timer request ID?  For example, say I'm implementing a routine toL >be called after mapping into memory via LIB$FIND_IMAGE_SYMBOL.  The routineO >needs to use a timer to determine if/when a timeout condition occurs.  There'stM >no guarantee that the calling routine/image isn't using any timers.  How canDO >my routine set a timer with a unique timer request ID, such that if cancelled,o; >it won't effect any timers from the calling routine/image?u  M I don't believe that there is any "supported" way to do this.  If you controlTL the calling code, you can create a routine which can manage your request IDsM but there is no guaranteed way otherwise.  I've often wished that there woulddO be a mechanism similar to the SCS SCS$ALLOC_RSPID routine to grant an identify-t ing value for such purposes.  n --O VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COMt            MO city, n., 1. a place where trees are cut down and streets are named after them.    ------------------------------  % Date: Tue, 24 Oct 2000 09:33:00 -0400y/ From: "Webb, William W" <wwebb1@email.usps.gov>r! Subject: eBay (guess what) again.eI Message-ID: <D46FE9B132FB9B44AEC242A96E4AB7502CB284@rlghncst625.usps.gov>c  L http://dailynews.yahoo.com/htx/cn/20001023/tc/ebay_blames_upgrade_for_system _troubles_1.html   Monday October 23 10:00 PM EDT( eBay blames upgrade for system troubles     By Troy Wolverton, CNET News.com  B eBay apologized to customers Monday for a system upgrade that has D led to intermittent outages and frustrating service problems during  the past two weeks.i  A  In a note to members on its annoucements board, eBay apologized  D for the problems and blamed them on an upgrade of its database that  it began in August.c  A "It may seem like it should take a short time to implement these sA improvements," eBay said in its note. "But it is quite a task to  B perform installations and testing while keeping the existing site : running. We conduct lots of tests to make sure that these A transitions are as seamless as possible. Sometimes they are not. V! And for that, we are very sorry."e  ? eBay representatives did not respond to repeated calls seeking rC comment. In its note, eBay said it should complete the upgrade the d next "seven to 10 days."  B The repeated system problems infuriated many auction sellers, who E took to the message boards on eBay, Honesty.com and AuctionWatch.com sE to complain. Sellers were upset that eBay refused to extend auctions   affected by the downtimes.  ? "I decided to stop selling until eBay fixes their problems and  D offers real customer support when things go wrong," one eBay member A said. "I used to come here to buy and sell because it's fun. I'm i9 getting absolutely no fun now and not much money either."u  ? The system upgrade and subsequent problems come amid a stellar 1B earnings report last week from the San Jose, Calif.-based company < that gave support to the company's sky-high future earnings < expectations. In a discussion of its financial results with @ investors and analysts, eBay executives said that the company's = investments in upgrading its technology had begun to pay off.u  B "We have made significant progress in stability and scalability," A company chief executive Meg Whitman said in a note to members on i@ the announcements board. "Our site again performed above the 99  percent level this quarter."  B As a result, eBay has begun to scale back spending on operations, B Whitman said. But the latest system problems could slow down that  effort.   B On Thursday, eBay took its site down for unscheduled maintenance. @ The site was down about 40 minutes, eBay said. On Tuesday, eBay A had an hour-long unplanned outage. Despite the hour-long outage,  7 eBay did not extend auctions affected by the downtime. h? On Oct. 15, eBay took its system down to perform "preventative PA maintenance." The company warned members of the downtime just 10 '@ minutes before taking the system down. Although the maintenance B was scheduled to last 10 minutes, eBay said system was down about  40 minutes.o  ? "We apologize for this necessary inconvenience, and appreciate e> your understanding," eBay said in a note on its announcements ' board concerning the Thursday downtime.t  @ eBay began to upgrade its system in August to determine whether @ it could handle future customer and auction growth. In its note B to members Monday, the company gave more details on the technical > changes, explaining that it has upgraded its database program  from Oracle 7.3.4 to Oracle 8i.   A As part of the upgrade, eBay has begun to divide its system into rA different databases and onto different systems. The goal is that aB if one feature on the system goes down, the rest of the site will % still be available, the company said.   A The upgrade immediately resulted in a series of sporadic outages  @ and glitches involving the personalized "My eBay" pages and the $ bidding history of individual items.  > The current problems also have affected the My eBay pages and > the bidding history, as well as certain search features. eBay ? explained in its note that it turned off those features during M> high traffic times to keep the rest of the eBay system up and $ running as it upgraded the database.  9 "While all features are important, we try to temporarily  > disable those that have the most significant impact to system A load and those that minimize impact to people's ability to find, m. view and bid on items," eBay said in its note.  > But by turning off those features, eBay has made it difficult > for bidders to find and bid on items, said Rosalinda Baldwin, = who monitors eBay's system problems as editor of The Auction  < Guild online newsletter. Additionally, the problems members = have encountered have been worse than eBay has acknowledged, c: she said, with the hard outages much longer than eBay has  stated in its postings.a  > "The problems have been endless," Baldwin said. "If you can't @ search and bid on an auction site, what good is the site doing?  None."  > The system problems come as the holiday shopping season draws 7 near. Last year, eBay became a popular gray market for  6 hard-to-find toys such as the Amazing Ally doll. That > phenomenon could repeat itself this year with sellers listing = items such as the PlayStation 2, which is in high demand due   to production cutbacks.e  7 The system problems could be an indication that eBay's -< listings are already beginning to climb, said Mark Gambale, = an analyst who covers the online auction industry for Gomez. u< "We're likely to see more problems as eBay gets closer to a  spike in listings," he said.  < But eBay said it hoped to have the problems contained soon. : "Our goal is to make these improvements in advance of the . holiday season," the company said in its note.  9 eBay's policy is to extend auctions and refund fees only  $ when outages last two or more hours.  < eBay said in its note that it would re-evaluate that policy.  = "Some of you have written to us asking us to extend auctions  < during these periods of feature unavailability," eBay said. 8 "In order to retrieve and change the information in the ; database, we would have to completely shut down the entire i  system for at least 30 minutes."  < But eBay should listen to its members, said Peter Marino, a = site operations analyst for Jupiter Communications. Although O: eBay's investment in technology will probably prevent the ? hours-long outages the site experienced last year, the company s= still needs to be sensitive to its customers, he said. Upset .: by problems like those eBay has had lately, members could @ turn to sites such as Yahoo or Amazon that offer more stability.  B "They don't have the luxury other sites do concerning downtimes," 7 Marino said. "They've got people 24 hours a day making -% transactions that mean a lot to them..  > "If the problems continue, the key is not going to be so much @ on technical side, but on how they handle their customers. They ? may have to address that: how are they are being accommodating - to what their customers want."   =================== The details of this article make eBay sound like the largest u" kludge presently known to mankind.  ! SEVEN to TEN DAYS for an upgrade?. Let's see Andrew spin THIS.J ================== William W. Webb  Sr. OpenVMS Systems Supporto DSCMF/OSS/MS, USPS
 (919)874-3043    ------------------------------  % Date: Tue, 24 Oct 2000 15:25:58 -0000t- From: wspencer@ap.nospam.org (Warren Spencer) % Subject: Re: eBay (guess what) again. / Message-ID: <svbag685d1flcc@news.supernews.com>   1 wwebb1@email.usps.gov (Webb, William W) wrote in  > <D46FE9B132FB9B44AEC242A96E4AB7502CB284@rlghncst625.usps.gov>:  
 -- snip --  B >As part of the upgrade, eBay has begun to divide its system into B >different databases and onto different systems. The goal is that C >if one feature on the system goes down, the rest of the site will t& >still be available, the company said. >t  
 -- snip --  J It appears they are partitioning the application to increase availability I and scalability.  If so, IMHO, it's a clumsy and course hack that likely -F would be unneccessary it the application was hosted on OpenVMS, where @ galaxy and clusters provide native partitioning and incremental  scalability.   ws   --  3 << What if there were no hypothetical questions? >>o   ------------------------------  # Date: Tue, 24 Oct 2000 12:13:47 GMTd From: gboado@nahuelsat.com.are Subject: Re: FMS and GKS) Message-ID: <8t3udn$uag$1@nnrp1.deja.com>o  - In article <39D60B9A.F59E7C7E@earthlink.net>,r:   "David J. Dachtera" <djesys.nospam@earthlink.net> wrote:  > gboado@nahuelsat.com.ar wrote: > >oF > > Hi. I need to transfer the FMS and GKS packages from one 3100 to aD > > MVII. I have the required licenses, but the installation kit was lost.x? > > Does anyone know which files do I have to copy and modify?.  > > * > > Sent via Deja.com http://www.deja.com/ > > Before you buy.s >eD > If you can tell me what versions you have, I can try to locate the) > distributions when I get back from L.A.y >b > -- > David J. Dachtera- > dba DJE Systems  > http://www.djesys.com/ >0< > Unofficial Affordable OpenVMS Home Page and Message Board:! > http://www.djesys.com/vms/soho/2 >4H > This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings > is to be expected. >nB > Feel free to exercise your rights of free speech and expression. >tH > However, attacks against individual posters, or groups of posters, are > strongly discouraged.   B Thank you very much. I've already solved the FMS problem (the mostE urgent). The problem is that we have an old system that was deliveredsE without the installation kits. I am trying to build a backup (just innG case the hardware stops working) by using some spares, but the softwaret* installation is proving to be a nightmare.    & Sent via Deja.com http://www.deja.com/ Before you buy.e   ------------------------------  % Date: Tue, 24 Oct 2000 02:06:22 -0400C2 From: rdeininger@mindspring.com (Robert Deininger)1 Subject: Re: FS microvax 3100-85, 128mb, 6gb, 8mmeL Message-ID: <rdeininger-2410000206220001@user-2ive68i.dialup.mindspring.com>  L In article <39F4D21F.E109E4B3@fp.co.nz>, Bill Eaton <fozzie@fp.co.nz> wrote:   > Bill Eaton wrote:c >  > > Available ex Auckland:C > > Microvax 3100-85, 128mb memory, 2x 1.3GB disks, 2x 2.0GB disks,oH > > vt420 console, TTi8510 tape (exabyte 8500 compatible).  VMS license.$ > > Bill Eaton email fozzie@fp.co.nz > $ > Following up own message, sorry :( > 4 > In answer to some questions, the full details are: >  > Microvax 3100 model 85,c > 128MB ECC memory,t > RRD42 CD-ROM,t > VT420 console, > 2 x RZ58 disks (1.3GB each),& > 2 x Generic SCSI disks (2.0GB each),! > VMS license is unlimited users,c= > Compaq expect a fee for registering transfer of the license?$ > (NZ$250 as at sometime last year),F > 8MM tape drive can be regarded as optional, capacity is about 4.5GB,1 > VMS version currently installed is VMS 5.5-2H4.: > 2 > If anyone has a notion of how much (if anything)* > this might be worth, please let me know!   What a coinky-dink...0D     http://cgi.ebay.com/aw-cgi/eBayISAPI.dll?ViewItem&item=476477815  G I suspect the presence of tranferable licences will mean a big increaseV in the system's value.   -- f Robert Deininger rdeininger@mindspring.com4   ------------------------------  % Date: Tue, 24 Oct 2000 16:43:45 -0200 ) From: fabio_compaq@ep-bc.petrobras.com.br   Subject: Future of VT, etc . . .L Message-ID: <OF1B330A3B.02B11D25-ON83256982.0065F0E5@ep-bc.petrobras.com.br>  H It=B4s just a curiosity:  Do you know if there=B4s a "Wabi"-like produc= t to run under OpenVMS ?tH Wabi was a Sun product  used to run Windows applications under Solaris = ?cH I installed  it a few years ago in  with MS Office (16 bits) and worked=  
 nice ! ! !  H I think it would be useful if the next generation of VT terminals will = be based on Java or Citrix/ ICA. H To have something similar to Windows desktop or even X-Windows but much=   more light.   & Forgive me if I am being heretic ! :-)     Regards,   FC=    ------------------------------  # Date: Tue, 24 Oct 2000 08:45:03 GMTn% From: A.Greig@virgin.net (Alan Greig)i Subject: Re: Galaxy doubt 2 Message-ID: <39f54ac0.1273414963@news.newsguy.com>  F On Mon, 23 Oct 2000 18:27:22 GMT, Alan E. Feldman <alan48@my-deja.com> wrote:    E >THERE'S NOTHING WRONG WITH SPLITTING AN INFINITIVE. Check the FAQ at  >www.grammarlady.com, e.g. >c	 >Example:: > . >...to boldly go where no man has gone before.  D In fact the (London) Times writing style guide quotes "to boldly go"@ and credits Gene Rodenberry with bringing acceptability to splitD infinitives. It's usually just Latin purists that get upset - If youA couldn't split an infinitive in Latin then you shouldn't do it inu English.  H >The alternatives sound ridiculous. Putting boldly between "to" and "go"D >puts the emphasis on it. Similarly, "to go boldly" puts emphasis on >"go". >  >--I >Disclaimer: JMHO3 >Alan E. Feldman >alan48  &-) >dellnet.com >  >t' >Sent via Deja.com http://www.deja.com/d >Before you buy.   --
 Alan Greig   ------------------------------  % Date: Tue, 24 Oct 2000 08:35:02 -0500l1 From: "Dave Gudewicz" <david.gudewicz@abbott.com>  Subject: Re: Galaxy doubt08 Message-ID: <8t42v3$pe7$1@fizban.fizban.pprd.abbott.com>  J Not to belabor this, but somehow "to go boldly" doesn't have the same ringL to it as "to boldly go" does.  But you do grant poetic license to the phraseJ "to boldly go" and for that I and most of Star Trek fandom thank you.  ;-)     Dave...w  E "Charlie Hammond" <hammond@not@peek.ppb.cpqcorp.net> wrote in messagem0 news:8t24o9$pko$1@mailint03.im.hou.compaq.com... >i+ > In article <8t1vtv$crg$1@nnrp1.deja.com>,a. > Alan E. Feldman <alan48@my-deja.com> writes: > ..G > >THERE'S NOTHING WRONG WITH SPLITTING AN INFINITIVE. Check the FAQ at  > >www.grammarlady.com, e.g. > >  > >Example:n > > 0 > >...to boldly go where no man has gone before. > >.J > >The alternatives sound ridiculous. Putting boldly between "to" and "go"F > >puts the emphasis on it. Similarly, "to go boldly" puts emphasis on > >"go". >rD > IMHO splitting an infinitive should be avoided when it can be done > without becoming too awkward.w >dG > In the case in point, if you want to emphasise "boldy" you should saysF > "boldy to go".  Personally, I think that "to go boldly" reads betterG > than "to boldy go" -- but in the overal context  "to boldly go" scansmI > better, so I grant poetic license.  But none of the three is bad enoughxK > to be called a syntax error -- more a matter of opinion.  Writing is moree > art than science.  >p > --I >     Charlie Hammond -- Compaq Computer Corporation -- Pompano Beach  FLo USAAH >          (hammond@peek.ppb.cpqcorp.net -- remove "@not" when replying)L >       All opinions expressed are my own and not necessarily my employer's. >i   ------------------------------  % Date: Tue, 24 Oct 2000 11:04:33 +0100t/ From: Nigel Arnot <sysmgr@maxwell.ph.kcl.ac.uk>s& Subject: Re: Galaxy doubt (now way OT)6 Message-ID: <009F2132.A77D0128.3@maxwell.ph.kcl.ac.uk>  F > THERE'S NOTHING WRONG WITH SPLITTING AN INFINITIVE. Check the FAQ at > www.grammarlady.com, e.g.  > 
 > Example: > / > ...to boldly go where no man has gone before.. > I > The alternatives sound ridiculous. Putting boldly between "to" and "go"=E > puts the emphasis on it. Similarly, "to go boldly" puts emphasis on  > "go".9  ; But "boldly to go" does put the emphasis where it's wanted,cE and is also gramatically correct. I guess my English teacher would be 
 proud of me. a  C That said, it's a live language, and whatever English teachers say, H common usage seems to be moving towards acceptance of split infinitives,B and I can't think of a good reason to fight this particular trend.   	Yours,p
 		Nigel Arnot - 		NRA@MAXWELL.PH.KCL.AC.UK                   t  7 		"In the beginning there was nothing, which exploded."n   ------------------------------  # Date: Tue, 24 Oct 2000 16:35:02 GMT , From: alphaman-nix-spam@hsv.sungardtrust.com Subject: GnuPG status?) Message-ID: <8t4dnm$c96$1@nnrp1.deja.com>a   Hi,t  E Didn't someone recently post that they'd successfully ported GnuPG toeD OpenVMS?  Or was I dreaming?  Which would mean that I'm now dreaming aloud...  C I've searched Deja to no avail, and there's no mention of it in thel GnuPG developers' mail list.   TIA, Aaronh --< Aaron Sakovich                 SunGard Huntsville Operations< Make April 15th just another day:    http://www.fairtax.org/" "OpenVMS: the OS MS wanted NT 2b."    & Sent via Deja.com http://www.deja.com/ Before you buy.A   ------------------------------  % Date: Tue, 24 Oct 2000 10:15:01 -0400n0 From: Paul Anderson <paul.r.anderson@compaq.com>/ Subject: Re: HP4050 vs HP4050TN when using DCPSaC Message-ID: <paul.r.anderson-54526B.10150124102000@news.compaq.com>h  9 In article <009F2092.EA0C1BF0@SSRL04.SLAC.STANFORD.EDU>, l% winston@SSRL.SLAC.STANFORD.EDU wrote:   C > Now the people in a remote building have a new HP4050TN(?); they  G > swear to me that this just the same as the other 4050 except that it i > has a duplexing option.   I Are the two printers configured identically?  Please check to see if the dH PostScript/PCL sensing settings are set correctly.  You also might want A to see if both printers have the same revision level of firmware.r   Paul   -- p,    Paul Anderson, OpenVMS Engineering (DCPS),    Compaq Computer Corporation, Littleton MA   ------------------------------  % Date: Tue, 24 Oct 2000 13:37:01 -0400a) From: "Ebinger . Eric" <EEbinger@drc.com>o+ Subject: Inexpensive Laser Printers for VMSvB Message-ID: <7162F87E9EF4D311BA9900805FC1D3AE7A6225@and02.drc.com>  + I am planning to upgrade my home collection 3 of systems (AlphaPC164SX running Windows NT-Alpha,  9 DS10 and Digital Server 3305 running hobbyiest VMS 7.2-1) ; to include networking the systems (Twisted Pair) and adding = a laser printer.  The problem is that there don't seem to be e< any laser printers (supported by DCPS) that are in my price  range (>$1100).   ; Again, this is for my home systems, high performance is notj> one of the requirements, reliability is.  It is also important? that I be able to use the printer from any of the systems.  AnynA recommendations on inexpensive laser printers that work with DCPS D would be appreciated.  I would be particluarly interested in hearingD of any plans to support the HP2100TN printer as it seems the closestE fit (in price, networking, and inclusion of a postscript "emulator").    Thanks in advance.  p Eric Ebinger   ------------------------------    Date: 24 Oct 2000 09:51:03 -0500, From: koehler@eisner.decus.org (Bob Koehler) Subject: RE: Java on Openvms+ Message-ID: <XyQcR3ZRSV5j@eisner.decus.org>Z  | In article <910612C07BCAD1119AF40000F86AF0D8052848C0@kaoexc3.kao.cpqcorp.net>, "Main, Kerry" <Kerry.Main@compaq.com> writes:	 > Andrew,r > - > Perhaps you could explain something for me.s > H >>>>1.   There are a number of different versions of Java 1.2.x 1.3 etc. > FACT.  > I > So, are you saying that each new release of Java means a new Java Brando > standard is required?h >   E It helps to understand that Java releases are NOT upward compatable. cF Every release I've used was issued with a release of incompatabilities with the previous release.  C If you're running Java bytecode from an earlier compiler it may nott) perform the same function on a later JVM.-  = And of course, they're not downward compatable; can't add new0B features and be downward compatable.  So Java bytcode from a later. compiler may not run at all on an earlier JVM.  F IMHO, if you want something to run use a slightly earlier compiler andH deal with the small upward incompatability.  If you use Sun's latest you- can expect it to only run on a few platforms.p  F ----------------------------------------------------------------------? Bob Koehler                     | Computer Sciences Corporationb= NASA GSFC Flight Software       | Federal Sector, Civil GroupeE                                 | please remove ".aspm" when replyingp   ------------------------------  # Date: Tue, 24 Oct 2000 16:35:02 GMTh From: chucknichols@my-deja.com< Subject: Re: Linker performance on DS20 slower than VAX 6520) Message-ID: <8t4dnm$c95$1@nnrp1.deja.com>l  ) In article <39F00DE0.16B30089@gtech.com>,g@   Arne =?iso-8859-1?Q?Vajh=F8j?= <arne.vajhoej@gtech.com> wrote:! > chucknichols@my-deja.com wrote:aH > > We are in the process of converting a real-time application from VAXC > > VMS 5.5-2 to Alpha openVMS 7.2. We have purchased dual CPU DS20: AlphasD > > with 1GB memory and two 9 GB disks mirrored for the system disk, threei8 > > 9 GB disks RAIDed to give 18GB for application disk. > >eD > > Linking programs takes much longer on the Alpha, for example oneF > > program takes about 3 minutes to link on our VAX 6520 (128 Kb) but it( > > takes 12 - 13 minutes on the DS20!!! > >TE > > As part of the conversion we borrowed a model 800 (400 Mhz) Alpha  VMS = > > 7.1 and I did not get complaints about the linking times. 
 UnfortunatelyIB > > we did not actually check the times but all the users feel the linkingS  > > performance was good on 800. > > D > > Is there any basic tuning required for an out of the box VMS 7.2E > > system. I have checked the linking performance when I am the onlyg user > > and it is still bad. >G' > Linking is a very IO heavy operation.i >nH > First thing to check is disk fragmentation. I have seen x6 performance- > increases of linking after defragmentation.s >M > Arne >  Thank you for the information.  ? I defrag'd the disk and got some improvement but not much. I amRD wondering if the RAIDed disks are slowing down the IO. My theory was, that new disks would be much faster that oldA ones and that even if there was a penalty for RAIDing it would ber balanced by the faster speed.d   Chuckl    & Sent via Deja.com http://www.deja.com/ Before you buy.)   ------------------------------  , Date: Tue, 24 Oct 2000 12:12:49 +0200 (CEST): From: "Gotfryd Smolik, VMS lists" <gotfryd@stanpol.com.pl>5 Subject: Re: loss of linefeeds when typing a files-11rJ Message-ID: <Pine.LNX.4.21.0010241141000.25612-100000@irys.stanpol.com.pl>    Hello !  3 On Thu, 19 Oct 2000 salvatorelli@my-deja.com wrote:o [...] # +This is the service configuration:h [...]n! +and this is the hnmlog.com file:e' +SYSTEM> type dsa1:[smsc.log]hnmlog.com  +$!q +$ def/user sys$output sys$net$ +$ type/cont dsa1:[smsc.log]smsc.log +$!o [...]a  =  Excuse the long time for answer - but I am something busy...l1  Say me, if TYPE (without /CONT) works properly ?,;  I will re-check your setup (at least for curiousity !) buta probably not this week :( 0  Here must be a resolution, check with the .COM:  % $ pipe crlf="" ; crlf[0,16]=13*256+10o- $ open/share=write in dsa1:[smsc.log]smsc.log9 $ open/write out sys$net $ptr:- $ read/err=errInp in lin $ write out lin,crlf
 $ goto ptr   $errInp: $ stat=$status $ wait 0::'p2'" $ if stat.eq.%X1827A then goto ptr< $ sh sym stat ! for diagnostic of unknown/unexcpected errors' $ goto ptr	! or run to close the server1  C  The additional crlf in "write" is for correction of missbehaviour:e= IMHO - still - somewhere must be a bug... the code is suposedeB  correct it, but the service itself probably skips the record end.   +Thanks for your help. +     Marilena  > +  "Gotfryd Smolik, VMS lists" <gotfryd@stanpol.com.pl> wrote:+ [please insert citates *before* a answer !]e  6 +> On Mon, 16 Oct 2000 tony_barratt@my-deja.com wrote: [...]t +> +TIAi +> + +> +Tony  :  I am not wrong, that here are two differrent person ? -:)    Regards - Gotfryd -- DE =====================================================================gF $ ON F$ERROR("LANGUAGE","ENGLISH","IN_MESSAGE").GT.F$ERROR("NORMAL") - 		THEN EXCUSE/OBJECT=MET. $!                        GS@stanpol.zabrze.plE =====================================================================e   ------------------------------   Date: 24 Oct 2000 12:24:41 GMT3 From: gartmann@immunbio.mpg.de (Christoph Gartmann) - Subject: mail$system_flags, DECNET IV & aliasp0 Message-ID: <8t3v29$8sg$1@n.ruf.uni-freiburg.de>   Hello,  J VMS 6.2 on a MicroVAX 3100. According to the docs it should be possible toN suppress the DECnet node portion of an e-mail address in a homomgenous clusterM by defining MAIL$SYSTEM_FLAGS in the correct way. I assigned it a value of 7.jK Nevertheless all mail sent from this node contains still a DECnet nodename.2H In NCP I defined an alias name for the VAXen in the cluster so that theyK use this alias name and not their nodename. But shouldn't it be possible to0C prevent the nodename from appearing in e-mail messages alltogether?o   Regards,    Christoph Gartmann   H -----------------------------------------------------------------------+H | Max-Planck-Institut fuer      Phone   : +49-761-5108-464   Fax: -452 |H | Immunbiologie                                                        |H | Postfach 1169                 Internet: gartmann@immunbio.mpg.de     |H | D-79011  Freiburg, FRG                                               |H +--------- http://www.immunbio.mpg.de/home/english/menue.html ---------+   ------------------------------   Date: 24 Oct 2000 16:00:53 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)1 Subject: Re: mail$system_flags, DECNET IV & aliast6 Message-ID: <8t4bnl$92q$2@mailint03.im.hou.compaq.com>  f In article <8t3v29$8sg$1@n.ruf.uni-freiburg.de>, gartmann@immunbio.mpg.de (Christoph Gartmann) writes:  K :VMS 6.2 on a MicroVAX 3100. According to the docs it should be possible toeO :suppress the DECnet node portion of an e-mail address in a homomgenous clusterB2 :by defining MAIL$SYSTEM_FLAGS in the correct way.  B   Where in the documentation are you looking?  (Could this be someA   confusion over whether the DECnet transport or the more direct i3   local transport is used for intra-cluster email?)h  sM :                                                 I assigned it a value of 7.eL :Nevertheless all mail sent from this node contains still a DECnet nodename.I :In NCP I defined an alias name for the VAXen in the cluster so that theysL :use this alias name and not their nodename. But shouldn't it be possible toD :prevent the nodename from appearing in e-mail messages alltogether?  +   Exactly how is mail$system_flags defined?n  ,     $ define/exec/system mail$system_flags 7     On all nodes?n  E   I would expect that a received email message -- arriving via DECnetcH   and specifically via the Mail-11 protocol -- would contain either the G   originating node or (if set up for it) the originating cluster alias.eJ   I would tend to expect that the originating node would include a return K   email address, in other words.  (This is how local mail typically works,  >   even on a standalone -- though networked -- OpenVMS system.)  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  % Date: Tue, 24 Oct 2000 11:34:57 -0400y' From: Derek Konigsberg <konigd@rpi.edu>t# Subject: Need VMS on tape cartridgee' Message-ID: <39F5AC21.2A52D339@rpi.edu>e   Hello,D     We've got a VAXstation 4 with a TK70 tape drive that I'd like toC reinstall VMS on.  I think the drive takes the same cartridges as asH TK50, but I'm not completely sure.  Anyways, I would use the OpenVMS/VAXF hobbyist CD that I purchased, except for one big problem.  Any attemptC of mine to interface this DEC CD-ROM drive to the SCSI bus seems totF result in the blowing of fuses that cost $1.30 each time.  I'd like toG avoid interfacing anything to the machine besides the drives already inmB it, so tape install would be ideal.  The machine's previous OS wasC VAX/VMS 5.5-2.  Any version from that or newer would be acceptable. # I'll pay for the tape and shipping.b   Thanks,i     Derek Konigsberg     konigd@rpi.edu$     RPI Electronics Club (president)   ------------------------------  # Date: Tue, 24 Oct 2000 12:55:37 GMTl  From: jlsue <jlsuexxxz@home.com>& Subject: Re: Odd File Expiration Dates8 Message-ID: <jd1bvskdnu5f79f4pe78gv3nioluo15320@4ax.com>  5 Are you certain that nobody every issued the command i  " $  SET FILE/EXPIRATION_DATE=<DATE>  . They *can* explicitly set the expiration date.   jlsh  = On Fri, 20 Oct 2000 21:40:33 GMT, jbecker@ui.urban.org wrote:l  @ >Can you think of any reason other than the obvious one why fileC >expiration dates would be "too far" into the future? The "obvious"iE >reason is the documented behavior around the current system time ando >volume retention times. > E >There are a few funny circumstances that make me wonder if somethingi >else is going on. >'E >But first, the background facts. This is a non-clustered AlphaServer H >4100 5/533 running OpenVMS V6.2-1H3. For the last few years, disks haveG >had min & max retention settings of 30 days and 90 days, respectively; H >I've never seen an instance in which the settings were other than that.E >The only time anyone messed with the system time AFAIK was last yearBF >during a few hours of Y2K testing; no users were on the system at the( >time (controlled testing and all that). >pC >Today + 90 days is 18-JAN-2001, so I'm calling any expiration dateb% >beyond that a funny expiration date.  >e8 >The funny expiration dates fall into some odd patterns:H >- Scattered expiration dates occur throughout the period 19-JAN-2001 to" >24-FEB-2003 on about 5,000 files.A >- In 2004, a few hundred files have expiration dates from 16-FEBe >through 27-FEB.H >- About 10,000 files have expiration dates from 28-FEB-2004 through 27-D >MAR-2004 (including most dates in between instead of just scattered >dates).F >- Apparently without exception, the creation dates for the files with; >funny expiration dates are in the past and look plausible.>E >- All of the above dates occur in many different directories on mostiC >disks. An interesting exception is the disk that went on line mostaA >recently, a few months ago, which has no funny expiration dates.T > D >None of this seems to fit with someone just messing around with theF >system time or volume retention times, so that's why I'm wondering if! >there's something else going on.u >sF >I noticed the funny dates quite a few months ago (possibly last year,H >not sure), but I didn't pursue it at the time. I'm only wondering aboutH >it now because the funny dates are getting closer, which messes up someE >in-house system management tools that assume valid expiration dates.y3 >For example, previously, it was safe to use things C >like /EXPIRED /BEFORE="+90-" to ignore files with funny expirationo >dates.t >  >Any ideas?o      1 Not speaking for anyone, certainly not DEC/Compaqo- (get rid of the xxxx in my address to e-mail)n   ------------------------------  % Date: Tue, 24 Oct 2000 12:13:46 -0400e& From: "Mike Duffy" <mdduffy@erols.com>& Subject: Re: Odd File Expiration Dates+ Message-ID: <8t4cei$k3b$1@bob.news.rcn.net>c   No,n  ? It does not set them.  It recommends that you manually set thema3 on each volume upon which PerfectDisk is to be run.-  J It's to allow its internal concept of "warehouse, volatile, ordinary" etc.F files.  By looking at the combination of Minimum and Maximum retentionH on the volume and the expiration date of a file, it can get a rough idea@ (limited by the amount of difference between min and max) of theJ last time a file was read. If a file was written 10 years ago, but is readJ every day, PerfectDisk will know that it's not an inactive file because of the J expiration date, and will not mistakenly relegate the file to the furthest reaches of the disk.  @ PerfectDisk does _not_ manipulate any of these dates, but merely1 performs calculations using what's already there.g  7 The example given in the PerfectDisk doc gives a sample @ SET VOL/RET command.  If that example was followed at your site,D you will see the kinds of dates you're describing, about 5 1/2 years into the future.  9 Any Minimum and Maximum will work, but those dates closerr; together will allow more accurate calculations.  Too close, 9 and your system will be writing new expiration dates more-6 often than necessary.  That's why the example uses one# day difference between min and max..  = It is the system itself which rewrites expiration dates based$9 upon last file access and the current retention settings;t/ see $ HELP SET VOL /RET for the exact behavior.?  6 Please note that the system takes no action due to the9 presence of an "expired" file.  Thus, it's a handy way to - get an idea of the last time a file was read.l   -Mike Duffy     F jbecker@ui.urban.org wrote in message <8t1pcs$6po$1@nnrp1.deja.com>...F >We have indeed been running PerfectDisk. Does it set volume retentionC >dates, or otherwise do anything to give me the sorts of expiratione >dates I've been seeing? >-, >In article <8ssj8i$4v2$1@bob.news.rcn.net>,* >  "Mike Duffy" <mdduffy@erols.com> wrote:8 >> I should have mentioned this in my previous response: >>8 >> PerfectDisk does not really care what the Minimum and= >> Maximum retention periods are, it just uses the differenceA >> to do its calculation.@ >>; >> However, most sites do not use them, and the PerfectDiski6 >> documentation recommends enabling volume rentention7 >> dates and gives an example of a little over 5 years.  >> >> -Mike >> (A former Raxco employee) >>I >> jbecker@ui.urban.org wrote in message <8sqe4h$smi$1@nnrp1.deja.com>...iC >> >Can you think of any reason other than the obvious one why filepF >> >expiration dates would be "too far" into the future? The "obvious"H >> >reason is the documented behavior around the current system time and >> >volume retention times.- >> >	 >> [snip]  >> >> >, >--c >Jim Becker0, >The Urban Institute (http://www.urban.org/)8 >Encompass ESILUG (http://eisner.decus.org/lugs/esilug/) >- > ' >Sent via Deja.com http://www.deja.com/n >Before you buy.   ------------------------------    Date: 24 Oct 2000 10:34:29 +0200G From: Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de>t) Subject: Re: OpenVMS UK (London) RoadtriptH Message-ID: <y4wveybpt6.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de>  2 Didier Morandi <Didier.Morandi@Easynet.fr> writes:   > Hoff Hoffman wrote:cI > >   Members of Compaq OpenVMS Engineering will be in London for OpenVMSe > >   technical presentations:< > >     Enterprise House, High Holborn, Tuesday, 31-Oct-2000  3 Too bad - I will be at that place two days later...a   	Jan   ------------------------------  # Date: Tue, 24 Oct 2000 12:33:32 GMTI% From: A.Greig@virgin.net (Alan Greig)a) Subject: Re: OpenVMS UK (London) Roadtripp2 Message-ID: <39f58081.1287175730@news.newsguy.com>  @ On 23 Oct 2000 14:28:52 GMT, hoffman@xdelta.zko.dec.nospam (Hoff Hoffman) wrote:i   >  >-G >  Members of Compaq OpenVMS Engineering will be in London for OpenVMS o >  technical presentations:o >e# >    Enterprise House, High Holborns >    Tuesday, 31-Oct-2000r >cF >  Details and schedules for the OpenVMS technical update session and F >  on the other presentations to follow later -- likely via the DECUS   >  UK (www.decus.co.uk) website. > ) >  Having recently located my passport...a   Steve,  B I'm a bit confused: The Decus UK Newsletter lists 31st Oct and 1stB November but the included fax back form confirms 1st November. AreA there two distinct days with the second talking about the DII COEeE initiative as the paper docs seem to suggest or has the schedule just F been completely re-jigged. More confusingly the UK DECUS web site only lists the 31st.c  C Haven't been able to contact anyone at DECUS for clarification  yeta- and wouldn't want to tun up on the wrong day!w     --
 Alan Greig   ------------------------------  % Date: Tue, 24 Oct 2000 11:21:35 -0400tR From: jamese@beast.dtsw.army.mil (Ed James, 410-295-1919, ed.james@telecomsys.com)% Subject: Oracle 7 and 8i on same box?d0 Message-ID: <00102411213572@beast.dtsw.army.mil>   Hello,  G We are writing a new version of our OpenVMS Alpha based billing system.-D The current system uses Oracle 7.3.4.4, while we are writing the new version to use Oracle 8i.o  I Has anyone looked into running two databases using two versions of Oracler3 on the same OpenVMS machine? What were the results?a  E Is anyone actually running two databases using two versions of Oraclee on the same OpenVMS machine?  @ Any personal experience or appropriate URLs will be appreciated!  & Thank you very much for the help,   Ed  : Ed James                           ed.james@telecomsys.com5 TeleCommunications Systems, Inc.   voice 410-295-1919v; 2024 West Street, Suite 300              800-810-0827 x1919 5 Annapolis, MD 21401-3556           fax   410-280-1094l   ------------------------------  % Date: Tue, 24 Oct 2000 12:09:24 -0400l/ From: "McCarthy Kevin P." <McCarthyKP@BWSC.ORG> % Subject: Oracle 7 and 8i on same box? : Message-ID: <4C519CCC638BD411A4270000F8CD1D82034EA0@NTSV2>  H We have a two node cluster;  we are running 7.3.2.3.2 and 8.0.5.1 on oneF node and 7.3.3.6 on the other.  All of the oracle exes are on the sameE disk but are in different root directories ( [oracle7] , [oracle73] ,fE [oracle8] ).  You need to set your logicals properly before you start D each of the instances; no problems doing it though.  We will soon beH installing 8i on the second machine, it will then be running 7.3.3.6 and 8i.    Kevin McCarthy Boston Water & Sewer Commissione   -----Original Message-----9 From: jamese@beast.dtsw.army.mil (Ed James, 410-295-1919,l< ed.james@telecomsys.com) [mailto:jamese@beast.dtsw.army.mil]- Posted At: Tuesday, October 24, 2000 11:22 AMa Posted To: vms* Conversation: Oracle 7 and 8i on same box?% Subject: Oracle 7 and 8i on same box?f     Hello,  G We are writing a new version of our OpenVMS Alpha based billing system.eD The current system uses Oracle 7.3.4.4, while we are writing the new version to use Oracle 8i.l  B Has anyone looked into running two databases using two versions of Oracle3 on the same OpenVMS machine? What were the results?n  E Is anyone actually running two databases using two versions of Oraclej on the same OpenVMS machine?  @ Any personal experience or appropriate URLs will be appreciated!  & Thank you very much for the help,   Ed  : Ed James                           ed.james@telecomsys.com5 TeleCommunications Systems, Inc.   voice 410-295-1919P; 2024 West Street, Suite 300              800-810-0827 x1919 5 Annapolis, MD 21401-3556           fax   410-280-1094c   ------------------------------   Date: 24 Oct 2000 12:04:09 GMTF From: lederman@star.enet.dec.DISABLE-JUNK-EMAIL.com (Bart Z. Lederman) Subject: Re: PDF under OpenVMS* Message-ID: <8t3trp$577@usenet.pa.dec.com>   In article <381D0079763CB7C3.39B1E1CD3E96B16F.1CDA61D37A9CBCF3@lp.airnews.net>, Chris Scheers <chris@applied-synergy.com> writes:, >IH >As I understand it, Adobe's official PDF reader for VMS is the Java PDF >reader.  < As I understand it, there is no official PDF reader for VMS.< There is nothing on the Adobe web site which states that the9 Java-based PDF viewer is "official" or supported by Adobee on any platform.  = I would be very interested if anyone has a statement directlya! from Adobe that states otherwise.e   -- p(  B. Z. Lederman   Personal Opinions Only  8  Posting to a News group does NOT give anyone permission8  to send me advertising by E-mail or put me on a mailing  list of any kind.  5  Please remove the "DISABLE-JUNK-EMAIL" if you have ao5  legitimate reason to E-mail a response to this post.s   ------------------------------  # Date: Tue, 24 Oct 2000 12:48:43 GMT % From: A.Greig@virgin.net (Alan Greig). Subject: Re: PDF under OpenVMS2 Message-ID: <39f584b0.1288246540@news.newsguy.com>   On 24 Oct 2000 12:04:09 GMT,@ lederman@star.enet.dec.DISABLE-JUNK-EMAIL.com (Bart Z. Lederman) wrote:   >n >In article <381D0079763CB7C3.39B1E1CD3E96B16F.1CDA61D37A9CBCF3@lp.airnews.net>, Chris Scheers <chris@applied-synergy.com> writes: >>I >>As I understand it, Adobe's official PDF reader for VMS is the Java PDFT	 >>reader.  > = >As I understand it, there is no official PDF reader for VMS.== >There is nothing on the Adobe web site which states that the-: >Java-based PDF viewer is "official" or supported by Adobe >on any platform.h >t> >I would be very interested if anyone has a statement directly" >from Adobe that states otherwise.  @ Errr, there is such a statement on the adobe web site. Check out0 http://www.adobe.com/products/acrviewer/faq.html   Note the following paragraph:r    What is Adobe Acrobat Viewer?F Adobe Acrobat Viewer is a Java application that lets you view PortableF Document Format (PDF) files on any system where a Java Virtual MachineA (JVM) is installed. If you already have Adobe Acrobat Reader, youe@ won't need Acrobat Viewer. But if there is no version of AcrobatD Reader available for your system, and you have a JVM installed, thenE Acrobat Viewer is for you. Software developers can use Acrobat Viewer-E as an application or JavaBean. The Acrobat Viewer can be licensed foraA free to be bundled with Java-related products, or integrated into|F custom Java solutions through the JavaBean interface. Refer to the End- User License Agreement for more information. a  a   >-- ) > B. Z. Lederman   Personal Opinions Only  > 9 > Posting to a News group does NOT give anyone permission:9 > to send me advertising by E-mail or put me on a mailing  > list of any kind.- >-6 > Please remove the "DISABLE-JUNK-EMAIL" if you have a6 > legitimate reason to E-mail a response to this post. >m   --
 Alan Greig   ------------------------------   Date: 24 Oct 2000 15:59:31 GMTF From: lederman@star.enet.dec.DISABLE-JUNK-EMAIL.com (Bart Z. Lederman) Subject: Re: PDF under OpenVMS* Message-ID: <8t4bl3$9t8@usenet.pa.dec.com>  Z In article <39f584b0.1288246540@news.newsguy.com>, A.Greig@virgin.net (Alan Greig) writes: >>? >>I would be very interested if anyone has a statement directly # >>from Adobe that states otherwise.a >yA >Errr, there is such a statement on the adobe web site. Check oute1 >http://www.adobe.com/products/acrviewer/faq.html  >u >Note the following paragraph: >  > What is Adobe Acrobat Viewer? G >Adobe Acrobat Viewer is a Java application that lets you view Portable G >Document Format (PDF) files on any system where a Java Virtual MachineiB >(JVM) is installed. If you already have Adobe Acrobat Reader, youA >won't need Acrobat Viewer. But if there is no version of AcrobateE >Reader available for your system, and you have a JVM installed, thencF >Acrobat Viewer is for you. Software developers can use Acrobat ViewerF >as an application or JavaBean. The Acrobat Viewer can be licensed forB >free to be bundled with Java-related products, or integrated intoG >custom Java solutions through the JavaBean interface. Refer to the End . >User License Agreement for more information.  >    >--t >Alan Greig  >   @ Please point out where the word "supported" appears in the above, statement, or anywhere else on the web site.  ? When I last looked at it, the Java based viewer was distributed)> by Adobe without support.  I believe they even state that they  will not accept problem reports.  ? Again: if anyone has a statement from Adobe that clearly states-? that Adobe will support the Java based viewer, which means that0? they will accept problem reports and correct reported problems, ; or anything which clearly states that they will support the = viewer on OpenVMS, I would like to see it (and so would a loth of other people, I expect).    -- t(  B. Z. Lederman   Personal Opinions Only  8  Posting to a News group does NOT give anyone permission8  to send me advertising by E-mail or put me on a mailing  list of any kind.  5  Please remove the "DISABLE-JUNK-EMAIL" if you have at5  legitimate reason to E-mail a response to this post.s   ------------------------------  % Date: Tue, 24 Oct 2000 12:44:36 -0400 5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>t Subject: Re: PDF under OpenVMS, Message-ID: <8t4efm$2g4j$1@lead.zk3.dec.com>  I DPS has been removed from V7.3, it may have been still present in some FTgI software.  There should be "stubs" for the DPS client libraries that willcL allow images linked against DPS to run, but the calls to find/initialize theI extension should fail (as if you were trying to talk to a server that hadi DPS).D      . Brian Schenkenberger, VAXman- wrote in message' <009F1F92.347BDD10@SendSpamHere.ORG>...vF >In article <G2tFwL.3s@world.std.com>, moroney@world.std.spaamtrap.com (Michael Moroney) writes:i >>In articleJ <60333E9D00AE31E8.237882F2CA4681FE.216FA30F2142BBFB@lp.airnews.net>, Chris+ Scheers <chris@applied-synergy.com> writes:bJ >>> As I understand it, Display Postscript on VMS has been unsupported forI >>> some time, but if it was already installed, it would be left in placet! >>> during previous VMS upgrades.  >>>TD >>> This changes with VMS 7.3, which will remove Display Postscript. >>H >>Is it possible to "squirrel away" certain files and restore them after+ >>the upgrade to retain Display Postscript?  >2= >The E7.3 DECW server did seem to activate the DPS extension:i >s; >Activating extension image DECW$SVEXT_Adobe_DPS_Extension,e< >extension name: Adobe-DPS-Extension, entry address 00323DE0 >oD >However, a bug in the CDA viewer kept me from testing it.  I didn't3 >try pulling the older CDA viewer over to E7.3 yet.s >eB >Let's hope the "stubs" remain so we can continue to use the older >DPS extension files.  >  >--n3 >VAXman- OpenVMS APE certification number: AAA-0001o VAXman(at)TMESIS(dot)COM >rJ >city, n., 1. a place where trees are cut down and streets are named after them.g   ------------------------------  % Date: Tue, 24 Oct 2000 12:40:08 -040005 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>E Subject: Re: PDF under OpenVMS, Message-ID: <8t4e7a$2g43$1@lead.zk3.dec.com>  I I know this is hard for people to accept - but we have no choice.  It hassL already been done on Tru64 UNIX, and we have committed to Adobe to remove itL from OpenVMS in all future releases.  Our hands are completely tied on this,L we have no choice in the matter.  It is not simply a matter of support (thatE reference was in regard to prior versions of VMS/Motif already in the,H field), we no longer have the right to continue shipping the proprietaryI Adobe software  in new versions.  It is not a VMS conspiracy, as I said -o! this also happened to Tru64 UNIX.         K John Johnstone wrote in message <39EF853D.1D10CE5B@onay-amspay.mail.com>...q >Fred Kleinsorge wrote:s >> >[snip]aK >> To use DPS will require that you use both pre-V7.3 and pre-V1.2-6.  Even I >> then, if it doesn't work - you are pretty much SOL.  We no longer havei thesK >> ability to get support from Adobe, or to obtain objects built debug from I >> them.  I've already had to close several bug reports with a "sorry, wen can'te >> support or fix it". >dJ >Since leaving DPS in wouldn't leave us any worse off than we already are,L >does Compaq have any other reason for removing DPS?  I thought Adobe hasn'tH >given support to Compaq for DPS for quite a while.  It seems surprisingI >that the lack of support should be an issue now.  Is it just coincidence4I >that Apple is making such a big deal about MacOS X using DPS from Adobe?sH >Here I go with a conspiracy theory, but it really seems as if Apple andF >Adobe don't want DPS used anywhere else.  If so, that's spectacularlyD >annoying since no one is really asking for or expecting there to be> >continued support of DPS in VMS, just leaving it in as it is. >iL >Has DPS been incorporated in any other X servers outside of Compaq?  If so,  >is DPS being yanked there also? >.K >I just realized earlier today that if DPS goes away, DECwrite's ability to G >display linked-in EPS files goes away also.  That's a smaller hit thandI >losing CDA Viewer but it's still another pain in the butt.  Like so many.L >other things, since I'm using DECwrite on the cheap via the CSA license PAK3 >bundle, I guess that's just life in today's world.m > H >I know that losing DPS is affecting only a small number of users but itK >really is a major loss.  I just remembered that DEC Document's Rags editordJ >will stop working.  I'm sure it's not used all that often either but that >really stinks also.   ------------------------------  # Date: Tue, 24 Oct 2000 17:31:03 GMT = From: system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-)e Subject: Re: PDF under OpenVMS0 Message-ID: <009F2147.1E733019@SendSpamHere.ORG>  d In article <8t4e7a$2g43$1@lead.zk3.dec.com>, "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> writes:J >I know this is hard for people to accept - but we have no choice.  It hasM >already been done on Tru64 UNIX, and we have committed to Adobe to remove itxM >from OpenVMS in all future releases.  Our hands are completely tied on this,aM >we have no choice in the matter.  It is not simply a matter of support (that$F >reference was in regard to prior versions of VMS/Motif already in theI >field), we no longer have the right to continue shipping the proprietary J >Adobe software  in new versions.  It is not a VMS conspiracy, as I said -" >this also happened to Tru64 UNIX.  % No but perhaps a Billy conspiracy? ;)m  K I can see this cutting into the 'update' revenues for VMS.  I know now that J I will not be updating DECwindows on my boxes.  You might as well stop any DECwindows development now.a   --O VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COMi             O city, n., 1. a place where trees are cut down and streets are named after them.s   ------------------------------  % Date: Tue, 24 Oct 2000 16:50:03 -0200 ) From: fabio_compaq@ep-bc.petrobras.com.bro Subject: Re: PDF under OpenVMSL Message-ID: <OF25BC350F.43FE9CF3-ON83256982.00674E44@ep-bc.petrobras.com.br>  H If there=B4s something like "Windows" running under OVMS there=B4s no n= eed of X-Windows !sH If a "Windows"  can emulate VMS why not  OpenVMS emulating a "Windows" = :-)c     Regards,   FC.U              . system@SendSpamHere.ORG em 24/10/2000 15:31:03  ) Favor responder a system@SendSpamHere.ORGSH                                                                        =     =20-H                                                                        =     =20rH                                                                        =     =20g    @                                                              =20@                                                              =20@                                                              =20@  Para:    Info-VAX@Mvb.Saic.Com                              =20@                                                              =20@  cc:      (bcc: Fabio dos Santos Cardoso/E-P-BC/Contratada)  =20@                                                              =20@                                                              =20@                                                              =20@  Assunto: Re: PDF under OpenVMS                              =20@                                                              =20           =d    > In article <8t4e7a$2g43$1@lead.zk3.dec.com>, "Fred Kleinsorge"% <kleinsorge@star.zko.dec.com> writes:OH >I know this is hard for people to accept - but we have no choice.  It = has H >already been done on Tru64 UNIX, and we have committed to Adobe to rem= ove  itH >from OpenVMS in all future releases.  Our hands are completely tied on=   this,,H >we have no choice in the matter.  It is not simply a matter of support=   (thattF >reference was in regard to prior versions of VMS/Motif already in theH >field), we no longer have the right to continue shipping the proprieta= ryH >Adobe software  in new versions.  It is not a VMS conspiracy, as I sai= d -o" >this also happened to Tru64 UNIX.  % No but perhaps a Billy conspiracy? ;)o  H I can see this cutting into the 'update' revenues for VMS.  I know now = thatH I will not be updating DECwindows on my boxes.  You might as well stop = anyu DECwindows development now.t   --2 VAXman- OpenVMS APE certification number: AAA-0001 VAXman(at)TMESIS(dot)COM  H city, n., 1. a place where trees are cut down and streets are named aft= er them.i       =l   ------------------------------  % Date: Tue, 24 Oct 2000 08:57:13 -0400m, From: Jeff Schreiber <schreiber@process.com>0 Subject: Re: Press Release from Process Software/ Message-ID: <009F2120.DD668DF7.333@process.com>s  / JF Mezei <jfmezei.spamnot@videotron.ca> writes:M >aL >> In the agreement with Process, Innosoft will transition certain rights toO >> development and responsibility for support of PMDF products on all platformsc >> to Process Software.l > K >So Innosoft retains onwership of PMDF then ? Is this because Innosoft with  >focus solely on SIMS ?e >oO >What sort of freedom with Process have to develop PMDF into whatever they wantt5 >? Or will PMDF truly remain the base code for SIMS ?,  K     I think the best way to really explain this is by answering the freedom F     question.  To put it in a nutshell, we basically have an unlimitedM     permanent license to do what we want with the code.  Legally Sun needs toCJ     retain ownership of the code for the intent that they plan for it, and>     that is encorporating portions into their iPlanet product.  L     But all legal speak aside, PMDF is now part of the Process Software lineI     of products, and will be maintained, supported and developed as such.l  5                                                 -Jeff;   --/ Jeff Schreiber,            Process Software LLCi1 schreiber@mx.process.com   http://www.process.comg/    TCPware, MultiNet & PMDF: Stronger than Everr   ------------------------------  # Date: Tue, 24 Oct 2000 18:00:17 GMTs* From: "Merlin" <merlin@exoticambrosia.com>C Subject: Protect yourself and your computer get Evidence Eliminatore3 Message-ID: <P4kJ5.4475$mf1.36938@news.pacbell.net>v   Evidence Eliminator8  K  When you access the Internet, your computer keeps permanent hidden recordsiC  of your activity for months or years, including web sites visited,uH  documents, pictures, videos and sounds. This is a serious threat to theG  legitimate surfer's privacy. Evidence Eliminator is a solution to thisoH  problem. It analyzes and protects your whole hard drive and defeats all6  known Forensic Analysis Software.To download, Go Here4  http://www.evidence-eliminator.com/main.shtml?A1564   ------------------------------    Date: 23 Oct 2000 23:25:11 -07003 From: Eric Smith <eric-no-spam-for-me@brouhaha.com> " Subject: Re: Q: Why not (2^n)-bit?0 Message-ID: <qhn1fuycvs.fsf@ruckus.brouhaha.com>  ' A.Greig@virgin.net (Alan Greig) writes: E > I recall we had to fit resistors to all the serial lines to prevent F > front end reboots on long open lines. Otherwise the front end 11 was& > swamped responding to phantom input.  F Interesting approach, but the correct solution was to either use 20 mAF current loop, optically isolated RS-232, or short-haul modems.  If youI had RS-232 lines long enough to have the problems you describe, they weresE long enough to actually be a potential safety risk due to differencesaG in ground potential between the computer and the terminals.  RS-232 wasr- not intended for anywhere near that distance.n   ------------------------------    Date: 24 Oct 2000 10:12:16 +0200G From: Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de>v" Subject: Re: Q: Why not (2^n)-bit?H Message-ID: <y466mid5en.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de>  5 Eric Smith <eric-no-spam-for-me@brouhaha.com> writes:e  G > > I recall we had to fit resistors to all the serial lines to preventnH > > front end reboots on long open lines. Otherwise the front end 11 was( > > swamped responding to phantom input.H > Interesting approach, but the correct solution was to either use 20 mAH > current loop, optically isolated RS-232, or short-haul modems.  If youK > had RS-232 lines long enough to have the problems you describe, they were>G > long enough to actually be a potential safety risk due to differencesII > in ground potential between the computer and the terminals.  RS-232 was1/ > not intended for anywhere near that distance.s  J Yup. Had nice moments on the VAXen as well ("Why is this system respondingL sluggishly in the middle of the night?"). And then there were the UARTs that% needed replacing after thunderstorms.f   	Jan   ------------------------------  # Date: Tue, 24 Oct 2000 08:26:01 GMTa% From: A.Greig@virgin.net (Alan Greig)e" Subject: Re: Q: Why not (2^n)-bit?2 Message-ID: <39f5459f.1272101885@news.newsguy.com>  ) On 23 Oct 2000 23:25:11 -0700, Eric Smith.) <eric-no-spam-for-me@brouhaha.com> wrote:a  ( >A.Greig@virgin.net (Alan Greig) writes:F >> I recall we had to fit resistors to all the serial lines to preventG >> front end reboots on long open lines. Otherwise the front end 11 wasa' >> swamped responding to phantom input.  >FG >Interesting approach, but the correct solution was to either use 20 mAfG >current loop, optically isolated RS-232, or short-haul modems.  If youBJ >had RS-232 lines long enough to have the problems you describe, they wereF >long enough to actually be a potential safety risk due to differencesH >in ground potential between the computer and the terminals.  RS-232 was- >not intended for anywhere near that distanced  E Yes, I'm sure some of the lines definitely exceeded the RS232 spec. AhD few lines (such as those linking to the DEC-10 3/4 mile away did useE modems but the cost to was cosnsidered too high to do this across theoE board. If memory serves it was quite common for lines to snake across:= campus twisted around hot water pipes and the like in various C educational institutiions - making lightning strikes interesting...o  D The resistor mod (I think between TX/RX and GND but am not sure) was! recommended by DEC field service.-   --
 Alan Greig   ------------------------------  % Date: Tue, 24 Oct 2000 11:44:21 +0100I8 From: John Macallister <J.Macallister1@physics.ox.ac.uk>> Subject: RE: Question about running out of queue entry numbersL Message-ID: <35666012DF4CD411BE940090279FA240111F3E@ppnt41.physics.ox.ac.uk>  A >I can't recall nor can I imagine having, say 999 jobs waiting inr. >print/batch queues (possible, but unlikely),   J Anyone running SMTP software which queues mail via the VMS queueing systemE will be familiar with this on busy systems. It just takes an extended I network hiccup or even a local system hiccup to stop mail delivery and tooH generate thousands of queue entries. Although it could happen with batchH jobs or print jobs my guess is that it's more likely to happen with SMTPL mail nowadays. With systems handling 20,000 mail messages a day it's not tooI difficult to imagine the entry numbers reaching high values when problemsl occur.   John    B Name: John B. Macallister  E-mail: j.macallister1@physics.ox.ac.ukH Post: Nuclear and Astrophysics Laboratory, Keble Road, Oxford OX1 3RH,UKA Phone: +44-1865-273388 (direct)  273333 (reception)  273418 (Fax)o   ------------------------------   Date: 24 Oct 2000 11:07 -0400  From: hein@eps.zko.dec.c*m Subject: Re: Record not currentA& Message-ID: <24OCT200011072265@miasys>  T In article <8t2bb2$27a3$1@news.comnet.co.nz>, "kevinf" <kevinf@gns.cri.nz> writes...K >I am using ISAM files, performing a keyed read, deleting that record, theneJ >sequentially reading the next 'n' records and deleting them, too. UsuallyD >this is fine, but occasionally the delete trips up with "record not> >current", so far only on records that were read sequentially.  3 Let me guess... your programming language is Cobol?e  : Be sure to read the general rules on the delete statement.< depending on the ACCESS clause cobol will tell rms to delete; the currenbt record (SEQUENTIAL) or cobol will use the key t9 value fromt the current record buffers to (re-)access the 3 record with that key and then deletes it (DYNAMIC).m  H >Another process also accesses the same file with SHARED write access. IL >understand "next record" and "current record" pointers to be unique to each >running process,   C Correct. They are in fact unique to the particular access 'stream'.4B (a single process may have mutliple streams to a files from one or more shared opens).r  ; > so do not see how the other process should interfere witha% >the delete sequence mentioned above.e  @  In the ACCESS IS DYNAMIC case, there is a timing window where aF  concurrent user may have locked (or even deleted!) the target record.C  I seem to recall this a specifically a problem with duplicate keys D  and for applications that use the RMS RAB$V_WAT featuer that allowsB  them to WAIT for record lock and 'snatch' the records while it is  being re-locked.    >Any ideas from the group?    = If this did not help, be sure to follow up Hoff's suggesting >K gathering more data: software userd, exact errors and sub-errors and so on.t  
 Good Luck,   Hein.s   ------------------------------  # Date: Tue, 24 Oct 2000 15:35:47 GMTt From: thefantom@my-deja.comf6 Subject: Re: RSH Problem - Can someone try this for me) Message-ID: <8t4a8j$8qp$1@nnrp1.deja.com>   @ Thanks Mr Hoffman, the time you spent doing the research is muchG appriciated. I would like to send your findings to Microsoft but I felliA they would not have a clue on what the solution should be. I haveiF informed 'The Wizard' on the compaq site but he has not responded yet.   I wait in dispair.  C I run a program instead of the WAIT. The WAIT is really there as anS example of the problem.   A Maybe the solution is to download a third party version of RSH ori* REXEC, but I can't find any FREE versions.    6 In article <8s54hq$hkv$1@mailint03.im.hou.compaq.com>,&   hoffman@xdelta.zko.dec.nospam wrote: >t >iH >   In regard to an earlier report here from "thefantom@my-deja.com" (inE >   <8qcqtq$d98$1@nnrp1.deja.com>) of a problem issuing the followinguD >   command on a Microsoft Windows NT and (per local testing here atC >   Compaq) on a Microsoft Windows 2000 system targeting an OpenVMSr# >   system running TCP/IP Services:u >c. >     REXEC VaxNodeName -l username -n WAIT 10 >1B >   that will trigger the following error message on the Microsoft system:n >h- >     "Recv failed: Connection reset by peer", >dG >   One of the TCP/IP Services engineers has performed a protocol tracea and B >   has found that the Windows 2000 system correctly initiates the REXECtG >   session, and that OpenVMS properly responds.   At four minutes, ther< >   Windows 2000 system sends a FIN to OpenVMS on the syserr connection, F >   OpenVMS ACKs the FIN, responds with a FIN and sends a FIN on REXECD >   port 512, and the Windows 2000 system then responds with an RST. >XA >   Per the engineer, this sequence appears odd for the following- reasons: >oG >     o Generating the FIN at four minutes, and for no apparent reason.e >a9 >     o Responding with an RST, when the FIN is returned.  >dC >     o Displaying "connection reset by peer" when the local systema resett. >       the connection -- and not by the peer. >t> >   Based on the protocol trace, TCP/IP Services is responding
 appropriatelyo >   to the received FIN. >oC >   Please report this odd FIN behaviour to the folks at Microsoft.  >nH >  --------------------------- pure personal opinion ------------------- --------1 >    Hoff (Stephen) Hoffman   OpenVMS Engineeringi hoffman#xdelta.zko.dec.com >g >t    & Sent via Deja.com http://www.deja.com/ Before you buy.o   ------------------------------  # Date: Tue, 24 Oct 2000 16:31:12 GMT  From: gboado@nahuelsat.com.ar  Subject: Shared data) Message-ID: <8t4dgg$c4u$1@nnrp1.deja.com>n  E I am trying to rewrite the I/O modules of an application which sharesiG data through a shared image. I haven't the source code, but running theoF DISM32 tool I could analyze the assembly language generated by the DECF fortran compiler. How could I code in FORTRAN the access to this data?F (my FORTRAN knowledge is almost NIL). I tried with a COMMON block, butC it was compiled as static local data (I analyzed the code generatedaH using DISM32). I am running everything on a Vaxstation with OpenVMS 5.5.% Thank you very much for your support.h    & Sent via Deja.com http://www.deja.com/ Before you buy.l   ------------------------------  # Date: Tue, 24 Oct 2000 12:48:20 GMTk- From: "Richard D. Piccard" <piccard@ohio.edu>u5 Subject: Re: splitting infinitives (was Galaxy doubt)r( Message-ID: <39F58512.DD61A7F7@ohio.edu>  G There are a number of supposed English grammer "rules" that are at besttJ normative rather than descriptive.  They may be, or have been, descriptiveK of Latin, but simply are not true of English as hard and fast rules (thoughX4 they may accurately describe the more common usage).  K Splitting infinitives and ending sentences with prepositions both fall intom this category.  K Frankly, for a natural language, I believe that one can sensibly reject theaL concept of normative rules, except where they are used to resolve ambiguity.  "                                RDP     Nigel Arnot wrote:  H > > THERE'S NOTHING WRONG WITH SPLITTING AN INFINITIVE. Check the FAQ at > > www.grammarlady.com, e.g.m > >  > > Example: > >t1 > > ...to boldly go where no man has gone before.  > >oK > > The alternatives sound ridiculous. Putting boldly between "to" and "go" G > > puts the emphasis on it. Similarly, "to go boldly" puts emphasis onp	 > > "go".  >d= > But "boldly to go" does put the emphasis where it's wanted,yG > and is also gramatically correct. I guess my English teacher would be  > proud of me. >aE > That said, it's a live language, and whatever English teachers say,tJ > common usage seems to be moving towards acceptance of split infinitives,D > and I can't think of a good reason to fight this particular trend. >  >         Yours, >                 Nigel Arnotr* >                 NRA@MAXWELL.PH.KCL.AC.UK >-G >                 "In the beginning there was nothing, which exploded."c   --B ==================================================================B Dick Piccard                           Academic Technology ManagerB piccard@ohio.edu                                 Computer ServicesB http://oak.cats.ohiou.edu/~piccard/                Ohio University   ------------------------------  % Date: Tue, 24 Oct 2000 12:32:53 +0100n& From: Roy Omond <Roy.Omond@Compaq.com>+ Subject: Re: Stockholm Exchange OM and VMS? * Message-ID: <39F57365.907908B6@Compaq.com>   jlhm@my-deja.com wrote:   1 > The Saxess system is NOT running on OpenVMS, it + > use Sun servers, see www.omtechology.com. 3 > We, OM Technology, are investigating the cause of 3 > the problem and we don't known at this moment the  > exact reason for the problem.    Yikes ... Sun servers !    Static ?   Sorry, couldn't resist :-)  	 Roy Omond  Blue Bubble Ltd.- (certainly not squeaking on behalf of Compaq)n   ------------------------------  % Date: Tue, 24 Oct 2000 08:02:27 -0400 ( From: Jonas Lindholm <jlindholm@rcn.com>+ Subject: Re: Stockholm Exchange OM and VMS?P' Message-ID: <39F57A53.8C6DF856@rcn.com>     :-)  > The problem was lack of communication capacity to all members.@ Last freday was an extremly active day and that caused very high activity from all members.? So the Saxess system was unable to send back all information to . the members and started to queue as it should.B Because of the queueing to all the members it look like the system1 was down but instead was the information delayed.g8 The solution is to increase the bandwidh to the members.  7 If you can read swedish, please go to www.idg.se/cs fory$ a statement from Carl Johan Hgborn.   /Jonas Lindholmt: Still NOT an official statement from OM Group or any other
 OM companies.l   Roy Omond wrote:   > jlhm@my-deja.com wrote:h >a3 > > The Saxess system is NOT running on OpenVMS, itt- > > use Sun servers, see www.omtechology.com. 5 > > We, OM Technology, are investigating the cause ofe5 > > the problem and we don't known at this moment thes! > > exact reason for the problem.t >V > Yikes ... Sun servers !  > 
 > Static ? >e > Sorry, couldn't resist :-) >a > Roy Omondn > Blue Bubble Ltd./ > (certainly not squeaking on behalf of Compaq)    ------------------------------  , Date: Tue, 24 Oct 2000 11:24:45 +0200 (CEST): From: "Gotfryd Smolik, VMS lists" <gotfryd@stanpol.com.pl>5 Subject: Re: SUMMARY: installing VMS 7.2-1 on 433au ?mJ Message-ID: <Pine.LNX.4.21.0010241048490.25612-100000@irys.stanpol.com.pl>  - On Mon, 16 Oct 2000, David J. Dachtera wrote:-   +Michael Moroney wrote:  [...] M +> Also some "commodity" drives implement only enough of the SCSI protocol to F +> work with Windoze.  VMS will use SCSI commands that are "mandatory"G +> according to the SCSI spec but not issued by Windoze, so they may bet" +> unimplemented or not debugged.  +eC +Then Linux, *BSD, etc. SCSI drivers must be equally deficient, no?s  
  Probably.  H  Do any UN*X system allow block-mode queued (asynchronous & cancellable)E IO interface to user-mode programm ? For curiousity only, of course !w  C  Where UN*Xes can recover after remove_and_re-insert a system disk,h while system works ?  A  May be some SCSI command are usefull in the mentioned examples ?r   [...] G +Then how do you explain the fact that they work with almost everythinge +EXCEPT VMS?  G  If I am not wrong (correction welcome) here was point about a example:nC some disks doesn't properly answer for the "send physical geometry" F request (cyl/heads/sectors) - and the info was required for allocationE of copy of label block ! (AFAIR the new alghoritm provides some addi- B tional location based on prime numbers, but that was a example of  problem)>  And yes - haven't check if DOS can recover after a SCSI disk  is removed -;)    Regards - Gotfryd   --  E =====================================================================LF $ ON F$ERROR("LANGUAGE","ENGLISH","IN_MESSAGE").GT.F$ERROR("NORMAL") - 		THEN EXCUSE/OBJECT=MEn. $!                        GS@stanpol.zabrze.plE =====================================================================.   ------------------------------  # Date: Tue, 24 Oct 2000 07:34:20 GMTe/ From: Mike Price <mike.price@littlewoods.co.uk>n& Subject: Re: TCP/IP and load balancing) Message-ID: <8t3e1s$j07$1@nnrp1.deja.com>p  * In article <39F477A8.E5D416BA@Easynet.fr>,3   Didier Morandi <Didier.Morandi@Easynet.fr> wrote:c > SysAdmin wrote:i > >a > ../... > >  > > David J. Dachterat > H > I would love to know why Mike has TCP answers from the Famous Datchera > Esquire and not me... :-(t >s > D.   >  Maybe he just took pity on me??n     --B All opinions are my own and do not necessarily reflect those of my employer    & Sent via Deja.com http://www.deja.com/ Before you buy.3   ------------------------------  % Date: Tue, 24 Oct 2000 10:10:31 +0100m- From: Martin Walker <Martin.Walker@csf.co.uk>-& Subject: RE: TCP/IP and load balancingQ Message-ID: <2161381B2B8BD4118EFD006097D7C7A707610C@london_exch.london.csf.co.uk>   I Have recently upgraded a UCX v4.2 (I think) system to v5.0.  The customereH thought they had previously got load broker to work with v4.2 although IL think it had been done by Compaq / Digital support some years ago.  (WhetherH they'd be able to help now I doubt but you could ask).  Didn't bother to? understand the details as we had to go to 5.0 for v7.2 support.o  J If you go with the TCPIP v5 route, be aware that 5.0a is required and thatF the recent patch kit also affects Load Broker (we installed it using aG pre-release patch kit - I assume the changes we needed made it into the- kit).e  I IIRC, once we had round robin working OK, we just set the ROUND_ROBIN_OFFa% logical correctly and it just worked.Q   YMMV!     -----Original Message-----d8 From: 	Mike Price [mailto:mike.price@littlewoods.co.uk]  Sent:	23 October 2000 16:55@ To:	Info-VAX@Mvb.Saic.Com " Subject:	TCP/IP and load balancing  D I am starting to look at using load balancing and TCP/IP rather than LAT as we normally use. F I understand that it goes via DNS. - OK so the round robin bit is easy" but doesn't do any load balancing.G The UCX 5.0 (a) docs talk about the load broker that talk to the metrica server..G Unfortunatly we are still on 4.2 and I can't quite see what you need toeF do for that version. This is complicated by the fact that we are not aB pure VMS site (OK OK I know!!) so the DNS system runs on a Solaris system. A Can we still use the metric server on UCX 4.2 and get the SolarisuD system to talk to it - if so what are the correct UNIX buzz words toG say to get the right information (I have tried reading ths DNS and BINDTC books but can't see anyting about metric server and load balancing)-  C The way I read it we need to set up a VMS system as a secondary DNSiE server and get the other VMS system to use it as their DNS host. - is-= this right?? - if so I think upgrading to 5.0 would be easier:  D Finally, once we upgrade to 5.0a, can the load broker talk to a UnixG DNS system and update it the right way - it looks from the docs that ittD can but as I am not sure of the exact mechanism on the Unix I am not sure.e  G Can anyone help - sorry if this is a bit basic - I speak UCX but havn't   had to do this DNS stuff before.  ? pointers to manuals would be welcome if I have missed somethingl   As always  TIA     Mike --B All opinions are my own and do not necessarily reflect those of my employer    & Sent via Deja.com http://www.deja.com/ Before you buy.e  A This e-mail including any attachments is confidential and may be  F legally privileged. If you have received it in error please advise the@ sender immediately by return email and then delete it from your  system. B The unauthorised use, distribution, copying or alteration of this F email is strictly forbidden. If you need assistance please contact the& CSF help desk on (+44)(0)207 490 2727.   ------------------------------  % Date: Tue, 24 Oct 2000 06:11:48 -0500e+ From: "Main, Kerry" <Kerry.Main@compaq.com>t& Subject: RE: TCP/IP and load balancingN Message-ID: <910612C07BCAD1119AF40000F86AF0D8052848E5@kaoexc3.kao.cpqcorp.net>   Mike,m  I We have a Customer locally that used OpenVMS V6.2, UCX V4.1 and a OpenVMS-8 based DNS to do load balancing and it worked very well.   J They shut down systems with zero application impact by disabling logins onJ the system to be shutdown, and like LAT, that removed that system from theK list of systems to receive new connections. When all current connections to H that system had expired (users went home etc), then that system could beH shutdown - not even a notice to users that a system was going down. WhenK that system was rebooted, logins were re-enabled and that system then begane& taking on its share of the load again.  J The Customer is now using TCPIP V50A and OpenVMS V7.2, and are still using- the VMS DNS (stability, reliability etc etc )s  H Now, with respect to TCPIP V5.0A and load broker, you can implement thisJ with a non-VMS based DNS as long as that DNS is up to recent RFC spec's ..I mininum BIND V8.1.1 which allows dynamic updates. (TCP 5.0a is BIND 8.1.2  compliant I believe)   Here are a few pointers:K <http://www.openvms.compaq.com:8000/72final/6526/6526profile_contents.html>fL <http://www.openvms.compaq.com:8000/72final/6526/6526profile_contents_001.ht0 ml#toc_chapter_6> (single urls will likely wrap)   Regards,  
 Kerry Main Senior Consultantc Compaq Canada Inc. Professional Servicess Voice: 613-592-46607 Fax  :  819-772-7036 Email: Kerry.Main@Compaq.com     -----Original Message-----6 From: Mike Price [mailto:mike.price@littlewoods.co.uk] Sent: October 23, 2000 11:55 AMt To: Info-VAX@Mvb.Saic.Com " Subject: TCP/IP and load balancing    D I am starting to look at using load balancing and TCP/IP rather than LAT as we normally use.sF I understand that it goes via DNS. - OK so the round robin bit is easy" but doesn't do any load balancing.G The UCX 5.0 (a) docs talk about the load broker that talk to the metricn server.SG Unfortunatly we are still on 4.2 and I can't quite see what you need tooF do for that version. This is complicated by the fact that we are not aB pure VMS site (OK OK I know!!) so the DNS system runs on a Solaris system.hA Can we still use the metric server on UCX 4.2 and get the Solaris D system to talk to it - if so what are the correct UNIX buzz words toG say to get the right information (I have tried reading ths DNS and BINDtC books but can't see anyting about metric server and load balancing)   C The way I read it we need to set up a VMS system as a secondary DNS E server and get the other VMS system to use it as their DNS host. - iso= this right?? - if so I think upgrading to 5.0 would be easierw  D Finally, once we upgrade to 5.0a, can the load broker talk to a UnixG DNS system and update it the right way - it looks from the docs that itkD can but as I am not sure of the exact mechanism on the Unix I am not sure.t  G Can anyone help - sorry if this is a bit basic - I speak UCX but havn't-  had to do this DNS stuff before.  ? pointers to manuals would be welcome if I have missed somethingi   As always  TIA     Mike --B All opinions are my own and do not necessarily reflect those of my employer    & Sent via Deja.com http://www.deja.com/ Before you buy.5   ------------------------------    Date: 24 Oct 2000 10:27:02 +0200G From: Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de>h8 Subject: Re: Trade-offs regarding Installed Known ImagesH Message-ID: <y43dhmd4q1.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de>  " jlsue <jlsuexxxz@home.com> writes:  H > How, exactly, does /SHARE use more memory?  I believe that if GBLPAGESE > is set to 128,000, it only uses something like 8 bytes of memory (I C > may have done my math wrong here, so somebody will correct me I'mp< > sure... however, I know that it is a very little amount).   K On the VAX, it is definitly higher - global pages use S0, and that can be asM sparse resource, as well as requiring physical memory for the PTEs. The Alpha:I has changed considerably since V1.5, and then you can now put things intooN S2, and with three- (or four-, by now?) level page tables you can ply a lot of tricks.   H > I'm sure you aren't saying that /SHARE installs the image into memory *  - because that is definitely not correct.    No, I didn't want to imply that.  B > I've actually seen overall memory utilization go *down* by using
 > /SHARE.  [explanation snipped]r  F Of course. That's the whole point of sharing. My point was that if you  L >> A question to VMS engineering: do INSTALLs data structures scale properlyL >> when hundreds or thousands of images/sections are installed (e.g., use a * >> hash table instead of a linear search)?F > Wow, good question.  However, I'm not sure it matters.  If I run theF > image, I have to map the pages whether they're already in memory, orE > retrieved from disk.  I have a hard time seeing how a memory access 9 > would be slower than having each access read from disk.   K Nonetheless, every call to the image activator will search the KFE list and L every call to map an image section has to search the GST list. Remember thatN embarassing (to say the least) episode when somebody "modified" the debugger'sL symbol table insert algorithm to be O(n^2) instead of O(n log n)? Some debugM sessions reached the DBG> prompt after a few hours of elapsed and CPU time...cD (There is also that wonderful text "one day in the life of the imageF activator" or somesuch that decribes this for one of the PDP-11 OSes.)  F > As to one of the original questions... I don't think that installingF > images will reduce the open file count per process.  I am willing toD > be corrected in this, but you still must open and/or map the image7 > (and associated shareable images) to get at the code.u  L If it is installed /OPEN, the image activator doesn't have to open it. If itN is installed /SHARED, the image sections are already mapped. That is the whole' point (well, most of it), n'est-ce pas?   @ > I would install /open/share/header *any* image that is either:= > 1.	Shared by more than 2 people, and used quite a lot.  Or,e; > 2.	Not shared much (e.g., DIRECTORY, compilers, etc), buth& > 	executed quite often on the system.  G Taht's fine with me, subject to the condition that enough "free" memorys remains.   	Jan   ------------------------------  % Date: Tue, 24 Oct 2000 13:02:18 -0400 5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>o  Subject: Re: VAX to Alpha (DS10), Message-ID: <8t4fgt$2gih$1@lead.zk3.dec.com>  B 3D Labs seems to be trying not to make the chip the distinguishingI characteristic, as they move to being more board vendor than chip vendor.nC The VX1 actually uses the "P3" chip, which is I'm told functionally H identical to the "R3" (or even the same, it's all a little muddled, as IK think the R series also includes geometry setup for 3D or some such thing).        Ebinger . Eric wrote in messageg9 <7162F87E9EF4D311BA9900805FC1D3AE7A621D@and02.drc.com>...eH According to 3DLabs web site (http://www.3dlabs.com) the Oxygen VX1 usesG the "3Dlabs GLINT R3" graphics processor.  Aren't you glad you asked?t   Eric Ebinger   > -----Original Message-----F > From: Shane.F.Smith@Notes.FH.Com [mailto:Shane.F.Smith@Notes.FH.Com]* > Sent: Thursday, October 19, 2000 1:12 PM > To: Info-VAX@Mvb.Saic.Comr" > Subject: Re: VAX to Alpha (DS10) >S >T >GF > Would you care to share with us which graphics chipset the VX1 uses? >  > Shane  >t   ------------------------------    Date: 24 Oct 2000 17:36:58 +0200* From: eplan@kapsch.net (Peter LANGSTOEGER)* Subject: Re: Very long filenames in VMSTAR* Message-ID: <39f5ac9a$1@news.kapsch.co.at>  c In article <39EB0D45.30DB34E5@hrem.stm.tudelft.nl>, Jouk Jansen <joukj@hrem.stm.tudelft.nl> writes:  >Oswald Knoppers wrote:u >> Jouk Jansen wrote:wE >> > Trying to unpack the tar-ball of StarOffice, I got an error froma >> > VMSTAR-3.3-9: >> > tar: error creatingA >> > [JOUKJ.PUBLIC.OPENOFFICE.OO_605_SRC.API.COM.SUN.STAR.ADDRESS 5 >> > ]ADDRESSBOOKSOURCESIMPLEDATABASEFIELDMAPPING.IDLr >> > tar: error writing file= >> > [JOUKJ.PUBLIC.OPENOFFICE.OO_605_SRC.API.COM.SUN.STAR.ADDK9 >> > RESS]ADDRESSBOOKSOURCESIMPLEDATABASEFIELDMAPPING.IDL  >> n/ >> I think you are one directory level to deep.= >> =	 >> Oswald=2 >Tried it 2 levels higher but I'm getting the same >tar: error creating= >[JOUKJ.OO_605_SRC.API.COM.SUN.STAR.ADDRESS]ADDRESSBOOKSOURCE  >SIMPLEDATABASEFIELDMAPPING.IDLo >tar: error writing file9 >[JOUKJ.OO_605_SRC.API.COM.SUN.STAR.ADDRESS]ADDRESSBOOKSO-# >URCESIMPLEDATABASEFIELDMAPPING.IDL2 > F >This is what I expected since the depth of directories should be much >more on VMS7.2-1.  I No. ODS-2 keeps the same. Your dirlevel is too deep and the filenames are H too long for ODS-2. What OpenVMS Alpha V7.2[xxx] introduced is ODS-5 and0 as already written in this thread will help you.) ODS-2 can't help you, not even on V7.2-1.     F Hi, Andrew, did you suggest such long filenames to the developers justI to frustrate us VMS fans ? I so far have never seen a reason for looooong 5 filesnames (even without separation chars in them)...o   -- q< Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111-2651; Network and OpenVMS system manager  Fax.    +43 1 81111-888u< <<< KAPSCH AG  Wagenseilgasse 1     E-mail  eplan@kapsch.netH A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"   ------------------------------  % Date: Tue, 24 Oct 2000 09:49:52 -0400s% From: "Islandco" <sales@islandco.com>. Subject: Re: video card info at / Message-ID: <svb3ueej5kg45f@corp.supernews.com>    Zane  H Scotty corrected this issue - it is almost the same part number on these video cardsm  @ The (SN-)PBXGD-AD  IS the powerstorm 300 and is supported by VMS  G The Matrox Millenium II - from what I can find out is only supported onsG Linux and NT - this was corrected too... unless someone knows differents  K You would think with the popularity of VMS in the market that someone wouldaD be writing Video drivers for more "common cards" like the Matrox etc  I FYI - All - we are looking for a good source of cheap Permedia II Cards -t4 these seem to work fine with VMS (as entry level 3D)     Davide       -- Island Computers US Corporation  2700 Gregory Street 	 Suite 150n Savannah GA 31404  Tel: 912 447 6622  Fax: 912 201 0096t sales@islandco.com www.islandco.com  C This message and any files transmitted with it are confidential and J may be privileged and/or subject to the provisions of privacy legislation.H They are intended solely for the use of the individual or entity to whomE they are addressed. If the reader of this message is not the intendeds
 recipient,G please notify Island Computers US Corp immediately and then delete thisO message.I You are notified that reliance on, disclosure of, distribution or copying" of this message is prohibited.    ; Zane H. Healy <healyzh@shell1.aracnet.com> wrote in message / news:zo8I5.139$Qz2.37782@typhoon.aracnet.com...i. > Island Computers <sales@islandco.com> wrote: > > www.islandco.com/video.htm >  > David,J > Nice.  However, I've got a couple questions and a suggestion.  First theH > suggestion.  Having a column that tells if the card is 8-bit or 24-bit woulda > be *very* nice.r >eK > Now the questions.  Are you sure the 4D40 is VMS compatible?  I'd thoughtaL > that it wasn't.  Secondly, what does the 'Y+' next to the Matrox Millenium! > II card in the VMS column mean?l >n > Zane   ------------------------------  , Date: Tue, 24 Oct 2000 10:29:57 +0200 (CEST): From: "Gotfryd Smolik, VMS lists" <gotfryd@stanpol.com.pl> Subject: Re: VMS V7.2-1sJ Message-ID: <Pine.LNX.4.21.0010241028400.25612-100000@irys.stanpol.com.pl>  5 On Tue, 17 Oct 2000, Gotfryd Smolik, VMS lists wrote:  [...]s= + I can't say for sure (*probably* may answer next week), bute. +isn't the driver set on 7.2-1 not complete ??  9  Aghr... Haven't check what write: I have think of 7.1-2,) probably 7.2-1 *is* complete...     Regards - Gotfryd   --  E ======================================================================F $ ON F$ERROR("LANGUAGE","ENGLISH","IN_MESSAGE").GT.F$ERROR("NORMAL") - 		THEN EXCUSE/OBJECT=MET. $!                        GS@stanpol.zabrze.plE ======================================================================   ------------------------------    Date: 24 Oct 2000 19:04:32 +0200* From: eplan@kapsch.net (Peter LANGSTOEGER) Subject: Re: VMS V7.2-1<* Message-ID: <39f5c120$1@news.kapsch.co.at>   In article <Pine.LNX.4.21.0010241028400.25612-100000@irys.stanpol.com.pl>, "Gotfryd Smolik, VMS lists" <gotfryd@stanpol.com.pl> writes:i6 >On Tue, 17 Oct 2000, Gotfryd Smolik, VMS lists wrote: >[...]> >+ I can't say for sure (*probably* may answer next week), but/ >+isn't the driver set on 7.2-1 not complete ??  >u: > Aghr... Haven't check what write: I have think of 7.1-2,  >probably 7.2-1 *is* complete...  * It depends on the definition of "complete"+ There is a FIBRECHAN ECO for V7.2-1, too...p   -- d< Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111-2651; Network and OpenVMS system manager  Fax.    +43 1 81111-888 < <<< KAPSCH AG  Wagenseilgasse 1     E-mail  eplan@kapsch.netH A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"   ------------------------------  % Date: Tue, 24 Oct 2000 02:02:47 -040002 From: rdeininger@mindspring.com (Robert Deininger)! Subject: Re: vt520 music questionhL Message-ID: <rdeininger-2410000202480001@user-2ive68i.dialup.mindspring.com>  f In article <39F4DC87.93B9DA56@earthlink.net>, "David J. Dachtera" <djesys.nospam@earthlink.net> wrote:   > Antony Wardle wrote: > > 6 > > anyone know why when you press the shift button on= > > a vt520 five times you get a cute little three note scaleo: > > sound (going up) and when you press it five more times3 > > you get the same, but the notes are going down.) > > , > > It also draws a u shape and fills it in. > > ) > > I have no idea how i discovered this.n  0 A nervous habit?  That's how I bumped into it...  4 > > Haven't found anything else, and haven't noticed6 > > it on any other vt devices. Tried it on excursion,6 > > but windows popped up something about sticky keys. > D > That's exactly it! It's an accessibility feature for, for example,I > quadraplegics who can do little more than manipulate a wand clenched innH > their teeth. Obviously, such an example would make SHIFT+anykey rather& > difficult, if not for "sticky" keys.  D Actually, I've never seen a vt520.  But this is exactly the behaviorM you get on a Macintosh if a particular "accessibility" extension is installed G and activated.  Hmm.  I wonder if this is some standard, or if not, whoS stole the idea from who.  F It must be at least quasi-standard.  Antony's description matches what+ a Mac does too perfectly to be coincidence.    -- o Robert Deininger rdeininger@mindspring.comi   ------------------------------  % Date: Tue, 24 Oct 2000 08:41:30 +0800e4 From: Dave Sneddon - bigpond <dbsneddon@bigpond.com>! Subject: Re: vt520 music question + Message-ID: <39F4DABA.305CDCFC@bigpond.com>u   Antony Wardle wrote: > 4 > anyone know why when you press the shift button on; > a vt520 five times you get a cute little three note scaled8 > sound (going up) and when you press it five more times1 > you get the same, but the notes are going down.I > * > It also draws a u shape and fills it in. > ' > I have no idea how i discovered this.f > 2 > Haven't found anything else, and haven't noticed4 > it on any other vt devices. Tried it on excursion,4 > but windows popped up something about sticky keys. > 	 > Antony.-   From the manual...   "Accessibility AidD This feature allows a user with limited motor skills to use modifierE key combinations in a sequential manner rather than in a simultaneouscD manner.  All modifier key combinations are supported.  There are twoE operation states -- Latch and Lock.  The Latch state affects only the ? next key pressed.  When in the Lock state, all keys pressed arenD affected by the modifier until you press the same modifier key againD or press any other modifier key twice.  A small icon, indicating theF state, is displayed on the Keyboard Indicator Line or the Status Line.  - 	To enable:	Press any modifier key five timesc+ 	  Latch state:	Press any modifier key once + 	  Lock state:	Press any modifier key twice0: 	To disable:	Press and hold a modifier key while you press 			another key."   HTHe -- V Regards, Dave. I -------------------------------------------------------------------------yI David B Sneddon (dbs)  OpenVMS Systems Programmer   dbsneddon@bigpond.com I DBS software at ...   http://www.users.bigpond.com/dbsneddon/software.htmtI "Life is what happens to you while you're busy making other plans" Lennonh   ------------------------------  % Date: Tue, 24 Oct 2000 13:17:26 -0400t0 From: Paul Anderson <paul.r.anderson@compaq.com>' Subject: Re: What does DCPS really do ?tC Message-ID: <paul.r.anderson-A71E81.13172624102000@news.compaq.com>   6 In article <39F4A084.74106AFE@videotron.ca>, JF Mezei % <jfmezei.spamnot@videotron.ca> wrote:u  E > But in the case of n-UP printing (multiple pages on a single piece  E > of paper), doesn't DCPS fiddle with the showpage operator to cause bF > it scale and translate (and possibly rotate) instead of just asking $ > the printer to spit out the page ?  B Yes, but most of the fiddling causes extra work to be done by the  printer, not the host.  E > The reason I ask is that on a Microvax II, DCPS seems to be taking pF > large amounts of CPU time when I send postscript code that contains C > images. The serial line is not defined as "binary capable". So i eE > would assume that DCPS scans my postscript code to find any binary 31 > data and convert it to an acceptable encoding ?@  G DCPS does not scan your PostScript file and change it for any reason.  rG If the file is in binary PostScript, for example, it is left alone and b sent to the file that way.  F > If DCPS only "worked" during print setup and termination, shouldn't E > it simply copy data to the printer once the job is properly setup,  G > thus not requiring large amounts of CPU (on a uVAX-II, any amount of a > CPU is a large amount :-)t  I I share your pain.  I remember seeing large amounts of CPU taken by DCPS  D symbiont processes on MicroVAX systems too.  I couldn't say without I investigating exactly what's taking that time.  But, for example, if you OI print a document four-up, it will appear to take more time since there's u7 less waiting for the printer to physically print pages.t   Paul   -- t,    Paul Anderson, OpenVMS Engineering (DCPS),    Compaq Computer Corporation, Littleton MA   ------------------------------    Date: 24 Oct 2000 10:29:16 +0200G From: Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de>ML Subject: Re: What would be your ultimate dream VMS machine (desktop/tower) ?H Message-ID: <y4zojubq1v.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de>  " jlsue <jlsuexxxz@home.com> writes:  E > Lastly, they qualify the drives via extensive testing under lots oftN > workloads to work out any latent bugs. Yes, they can be more expensive [...]  I About three times as much for disks and RAM? Especially for the latter, Ii can't believe it.    	Jan   ------------------------------  + Date: Tue, 24 Oct 2000 12:40:16 +0000 (UTC)w' From: Osmo Kujala <kujala@tukki.jyu.fi>aL Subject: Re: What would be your ultimate dream VMS machine (desktop/tower) ?, Message-ID: <8t3vvg$bks$1@mordred.cc.jyu.fi>  ! jlsue <jlsuexxxz@home.com> wrote:y> > I would encourage you to attend the next DECUS/CETS/whateverF > conference, then.  There you should have ready access to many of theD > engineers who will be happy - or, at least there - to answer these > types of quesitons..  # Who's going to pay the tickets? :-)o  F > My experience is that they do quite a lot to modify and qualify SCSIF > (note, there's no STANDARD in SCSI) to ensure that data integrity is > maintaned.  B > Not to mention to ensure that certainly OS-specific features are > implemented correctly.  E > Lastly, they qualify the drives via extensive testing under lots of ( > workloads to work out any latent bugs.  G > Yes, they can be more expensive, but when the business depends on thehG > data (and it's my pager that gets called in the middle of the night),.H > I'd rather have something that works *better* than "most of the time".  C If you read my previous followup's you'd see that I appreciate thatyG engineering work. It's only too expensive for us. Good luck, that theret  are customers who can afford it.  J Our cluster works *better* than "most of the time". There has been NO suchC problems that DEC/Compaq qualified devices would have helped a bit.u  C VMS in our university (like in most others) is "doomed". I can keepA9 it only if I can run it cheaply (=almost free of charge).c  F > And if you're in an environment where the data isn't that important,< > why are they spending any $$$ on computing resources then?  I Can you try to understand difference between important and time critical?gG Our data is very important and it must be (and is) very safe in our VMSe system.d? $$$? Is $285/year for 4-5 node cluster (VAX&Alpha containing mys AStation500 "PC") much?c    ? When we have now agreed that there have to be tested, qualifiedrC OpenVMS supported device models, why not use them for OpenVMS "masssI market" too. Extra price of engineering work would then be insignificant.oG But in the mean time we must use third party devices and user groups or 5 some such could keep lists of "well behaving" models.s   regardse           Osmo Kujalaw   ------------------------------   Date: 24 Oct 2000 14:57:25 GMT2 From: mathog@seqaxp.bio.caltech.edu (David Mathog)L Subject: Re: What would be your ultimate dream VMS machine (desktop/tower) ?+ Message-ID: <8t480l$gi@gap.cco.caltech.edu>t   In article <y4zojubq1v.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de>, Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de> writes:# >jlsue <jlsuexxxz@home.com> writes:o >tF >> Lastly, they qualify the drives via extensive testing under lots ofO >> workloads to work out any latent bugs. Yes, they can be more expensive [...]e >TJ >About three times as much for disks and RAM? Especially for the latter, I >can't believe it. r >n  L What I find annoying is that they make all these claims about how they test G the drives etc, and yet there is no program or procedure provided that  D would let an end user perform an equivalent test.   Presumably this I software exists - I can't imagine the engineers are manually shifting the-H logic levels with jumpers - and it would be a very useful thing to have.   That is:  
 1.  shut downz$ 2.  plug in SCSI drive (unqualified)
 3.  reboot' 4.  $ddd:==$SYS$SYSTEM:DRIVEQUALIFY.exe>     $ddd/test=24:00:00 DKB200:  H If it passes the test the user accepts it, if it fails, he tries anotherF brand.  Plus if they did provide this it would become trivial for disk4 manufacturers to market drives as being "VMS ready".  F That we don't have this is a pretty clear indication that it's the 3X J price markup, rather than the disk reliability, which is being protected.    Regards,   David Mathog mathog@seqaxp.bio.caltech.edu7? Manager, sequence analysis facility, biology division, Caltech w   ------------------------------  % Date: Tue, 24 Oct 2000 13:04:29 -0400r2 From: rdeininger@mindspring.com (Robert Deininger)L Subject: Re: What would be your ultimate dream VMS machine (desktop/tower) ?L Message-ID: <rdeininger-2410001304290001@user-2ive7lp.dialup.mindspring.com>  P In article <8t480l$gi@gap.cco.caltech.edu>, mathog@seqaxp.bio.caltech.edu wrote:  N > What I find annoying is that they make all these claims about how they test I > the drives etc, and yet there is no program or procedure provided that sF > would let an end user perform an equivalent test.   Presumably this K > software exists - I can't imagine the engineers are manually shifting therJ > logic levels with jumpers - and it would be a very useful thing to have. > 
 > That is: >  > 1.  shut downo& > 2.  plug in SCSI drive (unqualified) > 3.  reboot) > 4.  $ddd:==$SYS$SYSTEM:DRIVEQUALIFY.exe   >     $ddd/test=24:00:00 DKB200: > J > If it passes the test the user accepts it, if it fails, he tries anotherH > brand.  Plus if they did provide this it would become trivial for disk6 > manufacturers to market drives as being "VMS ready". > H > That we don't have this is a pretty clear indication that it's the 3X L > price markup, rather than the disk reliability, which is being protected.    You think that's all the Compaq engineers do to qualify a device?  Just grab a few of each brand from the supermarket, and plug each one into the test-O-matic?ro Then order a batch of the one that works best, and print up new sticky labels while waiting for them to arrive?    If that's all they did, you might have a valid complaint about the price.  Maybe someone in this newsgroup should go into business selling VMS-qualified disk drives, with only half the Compaq markup, if it's so easy.   I suspect the reality for the Compaq engineers is somewhat more complex.  Doing all the tests a disk drive needs probably requires some specialized equipment that typical end users don't have, can't afford, and wouldn't know how to use.KIt's probably rare that ANY of the supermarket devices pass all the tests. I've always understood that DEC/Compaq modifies the most promising devices as needed to bring them up to spec.  I guess that often means improved device firmware, but I've heard of at least one case where circuit modifications were made on the logic board.i  O I think you have buried a whole lot of hard work in your simple-looking step 4.9L I don't believe "DRIVEQUALIFY" exists as a standalone program that could runJ on a single VMS machine and do what needs doing.  A first, simple example:M What good is the test of DKB200 if nothing else is happening on the SCSI bus?    Heck, I don't even think your step 2 is realistic!  What kind of cable/termination configuration are you testing with?  What, you didn't specify it?  A drive that passes your test on machine A might well fail on machine B, running all the same software.f   Now, I agree, it would be nice if Compaq supplied some utilities to help dig out some features of drives.  But the answers would only be approximate, for all the reasons raised in this thread.  P There is already an alternative available for "quasi-supported" VMS disk drives.HBuy bare, non-Compaq drives, but buy them from a reseller who works with VMS systems.  These folks will do your first level selection for you, and you should most often end up with something that works as well as your 4-step test would ensure.  You'll pay some markup, but nothing like the cost of fully-qualified Compaq drives.   --   Robert Deininger rdeininger@mindspring.comd   ------------------------------    Date: 24 Oct 2000 14:29:56 -05009 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen)eL Subject: Re: What would be your ultimate dream VMS machine (desktop/tower) ?+ Message-ID: <sULDLS6Tp$Do@eisner.decus.org>o  ` In article <8t480l$gi@gap.cco.caltech.edu>, mathog@seqaxp.bio.caltech.edu (David Mathog) writes: > In article <y4zojubq1v.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de>, Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de> writes:$ >>jlsue <jlsuexxxz@home.com> writes: >>G >>> Lastly, they qualify the drives via extensive testing under lots ofbP >>> workloads to work out any latent bugs. Yes, they can be more expensive [...] >>K >>About three times as much for disks and RAM? Especially for the latter, Ie >>can't believe it.  >> > N > What I find annoying is that they make all these claims about how they test I > the drives etc, and yet there is no program or procedure provided that l3 > would let an end user perform an equivalent test.-   David-  B I think to make this argument effectively it would be necessary toA cite the precise URL where CalTech gives away the course material H for those people who do not want to sign up for those expensive classes.   Larry Kilgallenn   ------------------------------   Date: 24 Oct 2000 12:35:34 GMT* From: helbig@astro.rug.nl (Phillip Helbig)D Subject: why just single-user licenses with the educational package?. Message-ID: <8t3vmm$3h8$1@info.service.rug.nl>  A Bob and Alice are students at a university.  Like many other such/E institutions, VMS was ditched years ago since "unix is cheaper".  The=E same dudes who said that VMS was bad and proprietary are now, just as-C folks have gotten used to unix and almost forgotten VMS, moving the@D people over to "industry standard" Windows systems.  Bad?  Certainly@ worse than VMS.  Proprietary?  Of course.  There are a couple ofH greybeards who have kept up with VMS; they are tolerated on campus sinceG some "mission critical" (read: we don't care if you lose a month's work D as long as it doesn't show up as a monetary loss in the bookkeeping)I applications are still running on some old VAXes.  A couple of years ago,rH the campus license agreement was terminated since individual license are% cheaper for the few VAXes still left.i  C Alice:  Bob, aren't you running VMS on that used ALPHA you got fromd e-bay?  
 Bob:  Yes.  G Alice:  I heard you're going to move out of academica and get a "properr job" after getting your Ph.D.   E Bob:  Right.  With the third child on the way, Sally and I need a bitA more job security.  " Alice:  What are you looking into?  @ Bob:  I hope to get a VMS job.  Fortunately, the new educationalD license from Compaq means that I could install and run VMS for free.C It's sure nice to be back---I could barely keep up with my hobbyisteC setup at home, what with the kids and all.  Nice that I can hone myt skills here.  H Alice:  That's great.  Mike, one of the old VMS guys here, is my cousin.E The subject of VMS came up in a conversation a few months ago, but he D seemed rather pessimistic.  Said something like "Digital should haveG marketed it better, but now it's too late".  Which seems strange, sinceuG I gather that Digital was bought by Compaq; surely marketing was better , when they were pushing just their own stuff?   Bob:  Unfortunately, no.   Alice:  Strange.  G Bob:  Yes, but stranger things have happened.  I'm hoping that this new F educational agreement will help VMS gain a foothold in academia again.A It's in everyone's interest: students get to work with a real OS,nD industry benefits since they can find enough VMS people to hire, and@ when folks like me move into industry, they will prompt more VMS% purchases---for real money this time.T  D Alice:  Sounds interesting.  Mike has encouraged me to look into VMSF from time to time, and from what he was saying, it seems worth lookingD into, but at the moment it would be too much for me to manage my ownF box.  Say, could you give me an account on your box?  It's on the LAN,! and I could just log in remotely.a   Bob:  Sure.t  . Alice:  Thanks....Hey, what's the matter, Bob?  F Bob [choosing at random a volume from the Grey Wall as tears well intoG his eyes]:  Shit, did they have to screw up AGAIN?!  This new agreement C only has single-user licenses.  I was thinking, great that the base.E licenses are included (which they weren't before), and I can just getCD unlimited user licenses from DECcampus---but that's been terminated.  > Alice:  That's a shame.  It's the places which have terminatedG DECcampus and places where they've never had it where they need to makeeC inroads, but with just single-user licenses, only folks who alreadyaD know how to manage their own box are going to benefit.  Certainly noF chance of winning new VMS folks.  It might be better than nothing, but? I really can't see why they didn't include multi-user licenses.  Bob, no!  G Bob holds the "DCL Dictionary" like a frisbee and says "I don't know ifsC this is going to hurt you, me, or Compaq more" and flings it at his B ALPHAstation, hitting the CD-ROM drive and causing the box to moveH across the table.  "It would be SO easy to make just a LITTLE change andH get A LOT of improvement.  And it wouldn't be a disadvantage to anyone."  G Alice:  If even you're giving up, I guess I had better forget about it.n  E Mike happens to be passing by and sticks his head in the door.  "Hey,r what's going on?"d  ) Alice:  Just a, err, hmm, "system crash".e  E Mike:  You just have to get used to that with Windows.  Now with VMS,  oh, why bother.s   ------------------------------  % Date: Tue, 24 Oct 2000 13:10:48 -0400 2 From: rdeininger@mindspring.com (Robert Deininger)H Subject: Re: why just single-user licenses with the educational package?L Message-ID: <rdeininger-2410001310490001@user-2ive7lp.dialup.mindspring.com>  I In article <8t3vmm$3h8$1@info.service.rug.nl>, helbig@astro.rug.nl wrote:e  / > Bob and Alice are students at a university.  t   Aren't Bob and Alice each supposed to get their own educational license, and install it where ever they need it?  Silly and inconvenient, but not insurmountable.a   -- , Robert Deininger rdeininger@mindspring.comr   ------------------------------   End of INFO-VAX 2000.595 ************************